SB@NL2014 abstract book (version dd

Transcription

Symposium booklet
Programme
Participants
Abstracts
Maastricht, The Netherlands, 2014 December 15 -16
SB@NL2014 symposium
SBNL2014_Symposium_Booklet_20141210v2.docx
CONTENTS & COMMITTEE
Contents & Committee
Symposium booklet
page 2 / 88
SB@NL2014 symposium
Symposium booklet
Table of contents
Contents & Committee ....................................................................... 2
Table of contents ................................................................................ 3
Symposium committee ....................................................................... 4
Programme ......................................................................................... 8
Attendance list .................................................................................. 11
Abstracts ........................................................................................... 27
Partners & Sponsors ......................................................................... 80
page 3 / 88
SB@NL2014 symposium
Symposium booklet
Symposium committee
Symposium chair:
Prof. Dr Ir Ilja Arts
Maastricht Centre for Systems Biology (MaCSBio) & Department of Epidemiology (EPID), Faculty
of Health, Medicine and Life Sciences (FHML),
Maastricht University (UM)
T: +31-43-388 2971
E: [email protected]
I: www.maastrichtuniversity.nl/web/Institutes/FHML/FHML/DepartmentsCAPHRI/Epidemiology.htm
SB@NL partner: MaCSBio (Maastricht).
Core members:
Prof. Dr Bert Groen
Quantitative Systems Biology, Department of Pediatrics, Centre for Liver, Digestive and Metabolic
Diseases,
University Medical Centre Groningen (UMCG)
T: +31-50-363 2669
I: www.rug.nl/fmns-research/systemsbiology/index
SB@NL partner: SBC-EMA (Groningen).
Prof. Dr Jaap Heringa
Centre for Integrative Bioinformatics (IBIVU), Faculty of Sciences (FEW) & Faculty of Earth and Life
Sciences (FALW),
VU University Amsterdam (VU)
T: +31-20-598 7649
I: www.ibi.vu.nl
SB@NL partner: NISB (Amsterdam).
Dr Ir Natal van Riel
Systems Biology/Computational Biology, Division of Biomedical Imaging & Modelling (BIOMIM),
Department of Biomedical Engineering (BME),
Eindhoven University of Technology (TUE)
T: +31-40-247 5506
I: bmi.bmt.tue.nl/sysbio
SB@NL partner: CBio (Eindhoven).
page 4 / 88
SB@NL2014 symposium
Symposium booklet
Diman van Rossum, PhD
SB@NL Secretary
Netherlands Systems Biology Platform (SB@NL)
c/o Centrum Wiskunde & Informatica
Postbus 94079, 1090GB Amsterdam
(Science Park 123, 1098XG Amsterdam)
T: +31-6-2260 7479
Prof. Dr Lodewyk Wessels
Bioinformatics & Statistics, Division of Molecular Biology,
Netherlands Cancer Institute (NKI)
T: +31-20-512 7987
I: bioinformatics.nki.nl
SB@NL partner: CSBC (Amsterdam).
Prof. Dr Hans Westerhoff
Synthetic Systems Biology (SSB) / Nuclear Organisation Group (NOG), Swammerdam Institute for
Life Sciences (SILS), Faculty of Science (FNWI)
University of Amsterdam (UvA)
T: +31-20-525 7930
I: www.science.uva.nl/sils/nog
page 5 / 88
SB@NL2014 symposium
Symposium booklet
Additional members:
Dr Ruud Boessen
Risk Analysis for Products In Development, Cluster Earth, Environment and Life Sciences,
Netherlands Organisation for Applied Scientific Research (TNO)
T: +31-88-866 3260
I: www.tno.nl/rapid
SB@NL partner: MSB-TNO (Zeist).
Prof. Dr Roeland Merks
Biomodelling and Biosytems Analysis Group, Life Sciences Group,
Centrum Wiskunde & Informatica (CWI)
&
Leiden University (LU)
T: +31-20-592 4117
I: biomodel.project.cwi.nl
Prof. Dr Jaap Molenaar
Biometris, Mathematics and Statistical Methods, Department of Plant Sciences,
Wageningen University & Research Centre (WUR)
T: +31-317-486 042
I: www.biometris.nl
SB@NL partner: WCSB (Wageningen).
Dr Richard Notebaart
Systems Biology Group, Centre for Systems Biology and Bioenergetics (CSBB) / Centre for
Molecular and Biomolecular Informatics (CMBI), Nijmegen Centre for Molecular Life Sciences,
Radboud University Nijmegen (RUN)
T: +31-24-361 9693
I: www.richardnotebaart.nl/science.html
SB@NL partner: CSBB (Nijmegen).
page 6 / 88
SB@NL2014 symposium
Symposium booklet
Prof. Dr Gert-Jan van Ommen
Department of Human Genetics, Centre for Medical Systems Biology,
Leiden University Medical Centre (LUMC)
T: +31-71-526 9401
I: www.cmsb.nl
SB@NL partner: CMSB (Leiden).
Dr Marijana Radonjic
Network Biology, SME for Bioinformatics Data Integration and Mining,
EdgeLeap B.V.
Utrecht
T: +31-6-5230 8940
I: www.edgeleap.com
Dr Wilfred Röling
Eco System Biology, Faculty of Earth and Life Sciences (FALW),
VU University Amsterdam (VU)
T: +31-20-598 7192
I: www.falw.vu.nl
Symposium secretariat:
Diman van Rossum, PhD
SB@NL Secretary
Netherlands Systems Biology Platform (SB@NL)
c/o Centrum Wiskunde & Informatica
Postbus 94079, 1090GB Amsterdam
(Science Park 123, 1098XG Amsterdam)
T: +31-6-2260 7479
Website
www.biosb.nl/sbnl2014
page 7 / 88
SB@NL2014 symposium
Symposium booklet
PROGRAMME
Programme
page 8 / 88
SB@NL2014 symposium
Symposium booklet
page 9 / 88
SB@NL2014 symposium
Symposium booklet
page 10 / 88
SB@NL2014 symposium
Symposium booklet
ATTENDANCE LIST
Attendance list
page 11 / 88
SB@NL2014 symposium
Symposium booklet
Dr Abudukelimu Abulikemu, PhD
Molecular Cell Physiology, VU University Amsterdam
De Boelelaan 1085, 1081HV Amsterdam, The Netherlands
[email protected]
Dr Michiel Adriaens, PhD
Academic Medical Center (AMC)
Meibergdreef 9, 1100DD Amsterdam, The Netherlands
[email protected]
Carme de Andrés
European Projects Office, ISCIII
C/ Sinesio Delgado, 4 (Entrada por Avenida Monforte de Lemos, 5), 28029 Madrid, Spain
[email protected]
Prof. Judith Armitage, PhD
Biochemistry, University of Oxford
South Parks Road, OX2 8BX Oxford, United Kingdom
[email protected]
Prof. Dr Ir Ilja Arts, PhD
Dept. of Epidemiology, Maastricht University
PO Box 616, 6200 MD Maastricht, The Netherlands
[email protected]
Ir Jasmijn Baaijens, MSc
Life sciences, Centrum Wiskunde & Informatica
Science Park 123, 1098 XG Amsterdam, The Netherlands
[email protected]
Prof. Gary Bader, PhD
The Donnelly Centre, University of Toronto
160 College St. #602, M5S3E1 Toronto, Canada
[email protected]
Prof. Rudi Balling, PhD
Luxembourg Centre for Systems Biomedicine (LCSB), University of Luxembourg
7, Avenue des hauts Fourneaux, L-4362 Esch-sur-Alzette, Luxembourg
[email protected]
Dr Matteo Barberis, PhD
Synthetic Systems Biology, SILS, University of Amsterdam
Sciencepark 904, 1098 XH Amsterdam, The Netherlands
[email protected]
Drs Farnaz Barneh
Proteomics research center, Shahid Beheshti University of Medical Sciences
Shaghayegh building, Kokab st. Golshahr BLVD. Karaj, 31379-14431 Tehran, Iran
[email protected]
Drs Nirupama Benis, MSc
Host Microbe Interactomics, Wageningen University & Research Centre
De Elst, 6708 WD Wageningen, The Netherlands
[email protected]
Mirjam van Bentum, BSc
SB@NL2014 symposium organisation
page 12 / 88
SB@NL2014 symposium
Symposium booklet
Mittenwalderstrasse 11, 11061 Berlin, Germany
[email protected]
Drs Sonja Boas, MSc
Life Sciences, Centrum Wiskunde & Informatica
Science Park 123, 1098XG Amsterdam, The Netherlands
[email protected]
Dr Muriel de Boer, PhD
Institute for Environmental Sciences, VU University Amsterdam
Rector Frederiklaan 8A, 1851 AE Heiloo, The Netherlands
[email protected]
Dr Wim P.H. de Boer, PhD
FALW, VU University Amsterdam
De Boelelaan 1105, 1081 HV Amsterdam, The Netherlands
[email protected]
Dr Ruud Boessen, PhD
Utrechtseweg 48, 3704 HE Zeist, The Netherlands
[email protected]
Drs Anne Boeter, MSc
Department science and innovation, Netherlands Organisation for Health Research and Development
(ZonMw)
Laan van Nieuw Oost-Indië 334, 2593 CE 's-Gravenhage, The Netherlands
[email protected]
Drs Anwesha Bohler, MSc
Department of Bioinformatics - BiGCaT, Maastricht University
Universiteitsingel 50, 6229 ER Maastricht, The Netherlands
[email protected]
Marc Jan Bonder, MSc
Genetics, University Medical Centre Groningen
Postbus 30.001, Genetica CB50, 9700RB Groningen, The Netherlands
[email protected]
Dr Gergana Bounova, PhD
Netherlands Cancer Institute (NKI)
Plesmanlaan 121, B0721, 1066 CX Amsterdam, The Netherlands
[email protected]
Dr Simone van Breda, PhD
Toxicogenomics, Maastricht University
Universiteitssingel 40, 6229ER Maastricht, The Netherlands
[email protected]
Benoit Carreres, MSc
SSB, Wageningen University & Research Centre
Groen van prinstererstraat 7, 6702C Wageningen, The Netherlands
[email protected]
Dr Rachel Cavill, PhD
PO Box 616, 6200MD Maastricht, The Netherlands
page 13 / 88
SB@NL2014 symposium
Symposium booklet
[email protected]
Kridsadakorn Chaichoompu, PhD
Montefiore Institute, University of Liege
Grande Traverse 10, Sart-Tilman, 4000 Liege, Belgium
[email protected]
Anastasia Chasapi, MSc
University of Lausanne
Rue de l'Industrie 3, 1005 Lausanne, Switzerland
[email protected]
Dr Susan Coort, PhD
Bioinformatics-BiGCaT, Maastricht University
Universiteitssingel 50, 6200MD Maastricht, The Netherlands
[email protected]
Miguel Correa Marrero, MSc
Bioinformatics, Wageningen University & Research Centre
Bornsesteeg 1, 15B 004, 6708GA Wageningen, The Netherlands
[email protected]
Jesse van Dam, MSc
Systems and Synthetic biology, Wageningen University & Research Centre
Dreijenplein 10, 6703 HB Wageningen, The Netherlands
[email protected]
Ir Mark Davids, MSc
Laboratory of Systems and Synthetic Biology, Wageningen University & Research Centre
[email protected]
Patrick Deelen, MSc
Genetics department, University Medical Centre Groningen
Postbus 30.001, Genetica CB50, 9700 RB Groningen, The Netherlands
[email protected]
Dr Kasper Derks, PhD
Genetics, Erasmus Medical Centre
Wytemaweg 80, 3015 CN Rotterdam, The Netherlands
[email protected]
Dr Rob Diemel, PhD
Netherlands Organisation for Health Research and Development (ZonMw)
Laan van Nieuw Oost Indie 339, 2593CE 's-Gravenhage, The Netherlands
[email protected]
Dr Aalt-Jan van Dijk, PhD
Biometris, Wageningen University & Research Centre
Droevendaalsesteeg 1, 6708PB Wageningen, The Netherlands
[email protected]
Dr Ellen Dirkx, PhD
Cardiology, Maastricht University
Universiteitssingel 50, 6200 MD Maastricht, The Netherlands
[email protected]
page 14 / 88
SB@NL2014 symposium
Symposium booklet
Prof. Roel van Driel, PhD
ISBE / University of Amsterdam
Dag Hammarskjoldlaan 12, 1902DX Castricum, The Netherlands
[email protected]
Dr Lars Eijssen, PhD
Bioinformatics-BiGCaT, Maastricht University
Universiteitssingel 50, 6229 ER Maastricht, The Netherlands
[email protected]
Daniel Ettema, BSc
Def Software
postbus 11554, 2502AN 's-Gravenhage, The Netherlands
[email protected]
Dr Karen van Eunen, PhD
Pediatrics, University Medical Centre Groningen
A. Deusinglaan 1, 9713 AV Groningen, The Netherlands
[email protected]
Prof. Dr Ir Chris Evelo, PhD
Bioinformatics - BiGCaT, Maastricht University
[email protected]
Ir Zandra Félix Garza, MSc
Biomedical Engineering, Eindhoven University of Technology
Hertog Hendrik van Brabantplein 1A, 5611 PD Eindhoven, The Netherlands
[email protected]
Ramouna Fouladi, MSc
Electrical Engineering, University of Liege
Grande Traverse 10, 4000 Liege, Belgium
[email protected]
Prof. Dr Lude Franke, PhD
Department of Genetics, University Medical Centre Groningen
P.O. Box 30001, 9700 RB Groningen, The Netherlands
[email protected]
JingYuan Fu
, Groningen, The Netherlands
Daniel Garza, MSc
Center for Molecular and Biomolecular Informatics, University Medical Centre St Radboud
Geert Grootplein 28, 6525 GA Nijmegen, The Netherlands
[email protected]
Dr Frank Glod, PhD
National Research Fund
6 rue Nic Biever, L1777 Luxembourg, Luxembourg
[email protected]
Prof. Jan van der Greef, PhD
Utrechtseweg 48, 3700 AJ Zeist Zeist, The Netherlands
[email protected]
page 15 / 88
SB@NL2014 symposium
Symposium booklet
Prof. Bert Groen, BSc
University Medical Centre Groningen
Gein Noord 51, 1391ha Abcoude, The Netherlands
[email protected]
Drs Angela Guerriero, MSc
Microbiology, Wageningen University & Research Centre
Dreijenplein 10, 6703HB Wageningen, The Netherlands
[email protected]
Dr Jurgen Haanstra, PhD
Department of Pediatrics, Center for Liver, Digestive and Metabolic Diseases, University Medical Centre
Groningen
, Groningen, The Netherlands
[email protected]
Ir Mark Hanemaaijer, MSc
Systems Bioinformatics / Molecular Cell Physiology, VU University Amsterdam
[email protected]
Prof. Jaap Heringa, PhD
Centre for Integrative Bioinformatics, VU University Amsterdam
De Boelelaan 1081a, 1081 HV Amsterdam, The Netherlands
[email protected]
Michiel Herpers
Systems and Synthetic Biology, Wageningen University & Research Centre
Rijnveste 45, 6708 PZ Wageningen, The Netherlands
[email protected]
Heide Hess
SystemsX.ch
Clausiusstrasse 45, 8092 Zurich, Switzerland
[email protected]
Dr Peter-Bram 't Hoen, PhD
Human genetics, Leiden University Medical Centre (LUMC)
Albinusdreef 2, 2300 RC Leiden, The Netherlands
[email protected]
Kevin Hof, BSc
Anjerhof 9, 6708EW Wageningen, The Netherlands
[email protected]
Dr Wim van den Hof, PhD
[email protected]
Prof. Stefan Hohmann, PhD
Chemistry and Molecular Biology, University of Gothenburg
Box 462, 40503 Gothenburg, Sweden
[email protected]
page 16 / 88
SB@NL2014 symposium
Symposium booklet
Prof. Rob Hooft, PhD
DTL Data, Dutch Techcentre for Life Sciences (DTL) / Netherlands eScience Center
Catharijnesingel 54, 3511 GC Utrecht, The Netherlands
[email protected]
Bastian Hornung, MSc
[email protected]
Drs Rob ter Horst, MSc
Systems Biology, CMBI, University Medical Centre St Radboud
PO Box 9101, 6500 HB Nijmegen, The Netherlands
[email protected]
Dr Aare Ignat, PhD
Eesti Teadusagentuur
A. Lauteri 5, 10114 Tallinn, Estonia
[email protected]
Ir Sultan Imangaliyev, MSc
Department of Preventive Dentistry, Academisch Centrum Tandheelkunde Amsterdam (ACTA)
Gustav Mahlerlaan 3004, 1081LA Amsterdam, The Netherlands
[email protected]
Ir Annika Jacobsen, MSc
VU University Amsterdam
Boelelaan, 1081HV Amsterdam, The Netherlands
[email protected]
Dr Mohieddin Jafari, PhD
Proteomics, Institut Pasteur of Iran (IPI)
69, Pasteur St., 13164 Tehran, Iran
[email protected]
Rajaram Kaliyaperumal, MSc
Department of Human Genetics, Leiden University Medical Centre (LUMC)
LUMC, Building 2, Einthovenweg 20, 2333 ZC Leiden, The Netherlands
[email protected]
Prof. Antoine van Kampen, PhD
Bioinformatics Laboratory, Academic Medical Center (AMC)
Meibergdreef 9, 1105AZ Amsterdam, The Netherlands
[email protected]
Dr Thomas Kelder, PhD
EdgeLeap
Hooghiemstraplein 15, 3514AX Utrecht, The Netherlands
[email protected]
Dr Rinke Klein Entink, PhD
[email protected]
Prof. Dr Jos Kleinjans, PhD
page 17 / 88
SB@NL2014 symposium
Symposium booklet
[email protected]
Ir Jan Bert van Klinken, PhD
Human Genetics, Leiden University Medical Centre (LUMC)
Einthovenweg 20, 2333 ZC Leiden, The Netherlands
[email protected]
Jasper Koehorst, MSc
[email protected]
Dr Theo de Kok, PhD
[email protected]
Walter de Koster, BSc
Haarweg 107-506, 6709 PW Wageningen, The Netherlands
[email protected]
Dr Marcus Krantz, PhD
Humboldt-Universität zu Berlin
Invalidenstrasse 42, 10115 Berlin, Germany
[email protected]
Marianne ter Kuile, BSc
Netherlands Organisation for Health Research and Development (ZonMw)
Laan van Nieuw Oost-Indië 334, 2593 CE 's-Gravenhage, The Netherlands
[email protected]
Esther Kuiper, MSc
Molecular Cell Biology, VU University Amsterdam
het Vergult Cabeltouw 12, 5211XN 's-Hertogenbosch, The Netherlands
[email protected]
Dr Jan Albert Kuivenhoven, PhD
Section Molecular Genetics, Pediatrics, University Medical Centre Groningen
ERIBA Building, Antonius Deusinglaan 1, 9713AV Groningen, The Netherlands
[email protected]
Dr Arnold Kuzniar, PhD
Genetics, Erasmus Medical Centre
Wytemaweg, 3015 CN Rotterdam, The Netherlands
[email protected]
Prof. Dr Laurens Landeweerd, PhD
Delft University of Technology / Radboud University Nijmegen
julianalaan 60, 2628bc Delft, The Netherlands
[email protected]
Dr Jan Lankelma, PhD
[email protected]
page 18 / 88
SB@NL2014 symposium
Symposium booklet
Drs Cor Lieftink, MSc
Division of molecular carcinogenesis, Netherlands Cancer Institute (NKI)
Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands
[email protected]
Dr Christian Linke, PhD
Synthetic Systems Biology, SILS, University of Amsterdam
Scienpark 904, 1098 XH Amsterdam, The Netherlands
[email protected]
Dr Elisa Liras, PhD
Monsanto Holland B.V.
Leeuwenhoekweg 52, 2661CZ Bergschenhoek, The Netherlands
[email protected]
Drs Xiaowen Lu, MSc
University Medical Centre St Radboud
Geert Grooteplein Zuid, 6525 GA Nijmegen, The Netherlands
[email protected]
Tjaša Marolt, BSc
Systems and synthetic biology, Wageningen University & Research Centre
Haarweg, 6709 Wageningen, The Netherlands
[email protected]
Ir Anne-Claire M.F. Martines, MSc
Antonius Deusinglaan 1, 9713 AV Groningen, The Netherlands
[email protected]
Prof. Roeland Merks, PhD
Life Sciences, Centrum Wiskunde & Informatica / Leiden University
[email protected]
Dr Klaus-Peter Michel, PhD
Methodological and Structural Development in the Life Sciences, ERASysAPP / BMBF
Kapelle-Ufer 1, 10117 Berlin, Germany
[email protected]
Prof. Dr Jaap Molenaar
Droevenedaalsesteeg 1, 6708 PB Wageningen, The Netherlands
[email protected]
Drs D.G.A. Mondeel, MSc
Synthetic systems biology, SILS, University of Amsterdam
Science Park 904, 1098 XH Amsterdam, The Netherlands
[email protected]
Drs Mona Moshayedi
Clinical pharmacy and therapeutics, Mazandaran University of Medical Sciences of Iran
NO.10, farshadi St. Salman Farsi sq. Mazandaran, Iran, 6803088901 Farshadi, Iran
[email protected]
Daniël Muysken
page 19 / 88
SB@NL2014 symposium
Symposium booklet
Hogeschool Leiden
kloekhorststraat 289, 1104 MP Amsterdam, The Netherlands
[email protected]
Prof. Dr Gerard Muyzer, PhD
Microbial Systems Ecology, University of Amsterdam
[email protected]
Dr Shintaro Nakayama, PhD
[email protected]
Dr Ekaterina Nevedomskaya, PhD
Molecular Pathology, Netherlands Cancer Institute (NKI)
Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands
[email protected]
Dr Frans van Nieuwpoort, PhD
ISBE / University of Amsterdam
Science Park 904, 1090GE Amsterdam, The Netherlands
[email protected]
Drs Niclas Nordholt, MSc
Systems Bioinformatics, Faculty of Earth and Lifesciences, VU University Amsterdam
[email protected]
Dr Richard Notebaart, PhD
Centre for Systems Biology and Bioenergetics, University Medical Centre St Radboud
Geert Grooteplein Zuid 10, 6500 HB Nijmegen, The Netherlands
[email protected]
Dorett Odoni, MSc
Wageningen University & Research Centre
[email protected]
Christian Oertlin, BSc
BiGCaT - Bioinformatics, Maastricht University
[email protected]
Drs Rurika Oka, MSc
University of Amsterdam
[email protected]
Prof. Dr Gert-Jan van Ommen
Humane Genetics, BBMRI-NL / Leiden University Medical Centre (LUMC)
P.O. Box 9600, 2300 RC Leiden, The Netherlands
[email protected]
Ir Henk van Ooijen, MSc
Precision & Decentralised Diagnostics, Philips Research
High Tech Campus, 5656AE Eindhoven, The Netherlands
page 20 / 88
SB@NL2014 symposium
Symposium booklet
[email protected]
Ir Lex Overmars, BSc
University of Amsterdam
POSTBUS 94248, 1090 GE Amsterdam, The Netherlands
[email protected]
Drs Yared Paalvast, MSc
University Medical Centre Groningen
Hanzeplein 1, 9700 RB Groningen, The Netherlands
[email protected]
Dr Nadine Peyriéras, PhD
CNRS
Avenue de la Terrasse, 91198 Gif-sur-Yvette, France
[email protected]
Diewertje Piebes, BSc
NOG SSB, University of Amsterdam
sciencepark 904, 1098 XH Amsterdam, The Netherlands
[email protected]
Drs Arne Poortinga
Institute of Life Science and Technology, Hanze University of Applied Science
Zernikeplein 11, 9749 AS Groningen, The Netherlands
[email protected]
Drs Iraes Rabbers, MSc
Systems Bioinformatics, VU University Amsterdam
[email protected]
Dr Marijana Radonjic, PhD
EdgeLeap
Hooghiemstraplein 15, 3514AX Utrecht, The Netherlands
[email protected]
Dr Samrina Rehman, PhD
School of Chemical Engineering and Analytical Sciences, Manchester Institute of Biotechnology, University
of Manchester
Princess Street 131, M1 7DN Manchester, United Kingdom
[email protected]
Maarten Reijnders, MSc
morfelden-walldorfplein 323, 6706LD Wageningen, The Netherlands
[email protected]
Prof. Dr Dirk-Jan Reijngoud, PhD
Internal ZIP code EA12, UMCG, PO Box 30.001, 9700RB Groningen, The Netherlands
[email protected]
Prof. Dr Ir Marcel Reinders, PhD
Pattern Recognition and Bioinformatics, Faculty of Electrical Engineering, Mathematics and Computer
Science (EWI), Delft University of Technology
Mekelweg 4, 2628CD Delft, The Netherlands
page 21 / 88
SB@NL2014 symposium
Symposium booklet
[email protected]
Ir Lisanne Rens, MSc
Life Sciences, Centrum Wiskunde & Informatica
[email protected]
Dr Natal van Riel, PhD
Den Dolech 2, 5612AZ Eindhoven, The Netherlands
[email protected]
Ir Rienk Rienksma, MSc
[email protected]
Teresa Robert Finestra, BSc
Sumatrastraat 12, 6707EG Wageningen, The Netherlands
[email protected]
Dr Wilfred Röling, PhD
Boelelaan 1085, 1081 HV Amsterdam, The Netherlands
[email protected]
Dr Diman van Rossum, PhD
SB@NL / BioSB research school
Touwbaan 40, 3142BV Maassluis, The Netherlands
[email protected]
Ir Yvonne Rozendaal, MSc
Den Dolech 2, 5612AZ Eindhoven, The Netherlands
[email protected]
Dr Edoardo Saccenti, PhD
[email protected]
Dr Luca Santuari, PhD
Plant Sciences Group, Wageningen University & Research Centre
Droevendaalsesteeg 1, 6700AP Wageningen, The Netherlands
[email protected]
Prof. Wim H.M. Saris, PhD
Human Biology, Maastricht University
[email protected]
Dr Peter Schaap, PhD
Laboratory of Systems and Synthetic biology, Wageningen University & Research Centre
Dreijeenplein 10, 6703HB Wageningen, The Netherlands
[email protected]
page 22 / 88
SB@NL2014 symposium
Symposium booklet
Dr Petra E. Schulte, PhD
Projectmanagement Jülich
Willhem-Johnen-Straße, 52428 Jülich, Germany
[email protected]
Prof. Benno Schwikowski, PhD
Institut Pasteur
25-28 rue du Dr. Roux, 75015 Paris, France
[email protected]
Drs Hanieh Shekarian, MSc
Hogeschool Leiden
Zernikedreef 11, 2333CK Leiden, The Netherlands
[email protected]
Ir Fianne Sips, MSc
Department of Biomedical Engineering, Eindhoven University of Technology
Den Dolech 2, 5612 AZ Eindhoven, The Netherlands
[email protected]
Bob van Sluijs, BSc
Dorpsstraat 107, 6871AE Renkum, The Netherlands
[email protected]
Bart Smeets, MSc
Bioinformatics, Maastricht University
[email protected]
Dr Robert Smith, PhD
Systems & Synthetic Biology, Wageningen University & Research Centre
Building 316, Dreijenplein 10, 6703HB Wageningen, The Netherlands
[email protected]
Dr Zita Soons, PhD
Department of Knowledge Engineering, Maastricht University
Postbus 616, 6200 MD Maastricht, The Netherlands
[email protected]
Terezinha de Souza, MSc
[email protected]
Dr Hans Stigter, BSc
Droevendaalsesteeg 1, 6866CB Wageningen, The Netherlands
[email protected]
Simona Stoian
UEFISCDI
D.I. Mendeleev, 21-25, 10362 Bucharest, Romania
[email protected]
Sarah Stolle, MSc
Systems Biology Centre for Energy Metabolism and Ageing, University Medical Centre Groningen
page 23 / 88
SB@NL2014 symposium
Symposium booklet
Antonius Deusinglaan 1, 9713AV Groningen, The Netherlands
[email protected]
Ir Nikolaos Strepis, MSc
Dreijenplein 4 Building 316 6703 HB, 6703 HB Wageningen, The Netherlands
[email protected]
Drs Rogier Stuger, MSc
[email protected]
Dr Maria Suarez-Diez, PhD
[email protected]
Drs Kalyanasundaram Subramanian, MSc
Leonard Roggeveenstraat 4, 6708SL Wageningen, The Netherlands
[email protected]
Drs Georg Summer, MSc
Cardiologie, Maastricht University
PO Box 616, 6200 MD Maastricht, The Netherlands
[email protected]
Drs Martina Summer-Kutmon, MSc
Department of Bioinformatics, Maastricht University
Universiteitssingel 50, Room 1.306, 6229 ER Maastricht, The Netherlands
[email protected]
Margit Suuroja, MSc
Estonian Research Council
Soola 8, 51013 Tartu, Estonia
[email protected]
Prof. Tsjalling Swierstra, PhD
Philosophy, Maastricht University
Grote Gracht 90-92, 6211SZ Maastricht, The Netherlands
[email protected]
Dr Radek Szklarczyk, PhD
Clinical Genomics, Maastricht University Medical Centre
[email protected]
Dr Hannan Tahir, PhD
Life Sciences Group, Centrum Wiskunde & Informatica
[email protected]
Ana Tardon, BSc
Dept. of International Research Programmes and External Relations, National Health Institute Carlos III
Av/Monforte de Lemos 5. Pabellón 6, 28029 Madrid, Spain
[email protected]
page 24 / 88
SB@NL2014 symposium
Symposium booklet
Prof. Bas Teusink, PhD
VU University Amsterdam
de Boelelaan 1085, 1081HV Amsterdam, The Netherlands
[email protected]
Drs Daniel Theunissen, MSc
Universiteitssingel 40 5.577B, 6229ER Maastricht, The Netherlands
[email protected]
Sebastian Thieme, MSc
Theoretical Biophysics, Humboldt-Universität zu Berlin
Invalidenstraße 42, 10115 Berlin, Germany
[email protected]
Bram Thijssen, MSc
Molecular Carcinogenesis, Netherlands Cancer Institute (NKI)
Plesmanlaan 121, 1066CX Amsterdam, The Netherlands
[email protected]
Prof. Kristel Van Steen, PhD
Electrical Engineering and Computer Science, University of Liege
Grande Traverse 10, B28, 4000 Liege, Belgium
[email protected]
Mira Vegter, MSc
Institute for Science, Innovation and Society, Radboud University Nijmegen
Molenstraat 41Bq, 6511HA Nijmegen, The Netherlands
[email protected]
Dr Malkhey Verma, PhD
School of Chemical Engineering, University of Manchester
Princess Street 131, M1 7DN Manchester, United Kingdom
[email protected]
Dr Pernette J. Verschure, PhD
Swammerdam Institute for Life Sciences, University of Amsterdam
[email protected]
Dr Bastienne Vriesendorp, PhD
R&D, Corbion
Arkelsedijk 46, 4206AC Gorinchem, The Netherlands
[email protected]
Agnieszka Wegrzyn, MSc
Antonius Deusinglaan 1, 9713AV Groningen, The Netherlands
[email protected]
Dr Ron Wehrens, PhD
Droevendaalsesteeg 1, 6708 PB Wageningen, The Netherlands
[email protected]
Prof. Dolf Weijers, PhD
page 25 / 88
SB@NL2014 symposium
Symposium booklet
Dreijenlaan 3, 6703HA Wageningen, The Netherlands
[email protected]
Prof. Lodewyk Wessels, PhD
Computational Cancer Biology, Netherlands Cancer Institute (NKI)
Plesmanlaan 121, 1066CX Amsterdam, The Netherlands
[email protected]
Prof. Hans V. Westerhoff, PhD
Synthetic Systems Biology, Molecular Cell Physiology and Integrative Systems Biology, University of
Amsterdam / Universities of Manchester
Manchester Centre for Integrative Systems Biology, Manchester Interdisciplinary Biocentre, 131 Princess
Street, Manchester , United Kingdom, EU, M1 7DN Manchester, United Kingdom
[email protected]
Prof. Ko Willems van Dijk, PhD
Human Genetics, Leiden University Medical Centre (LUMC)
Einthovenweg 20, 2333ZC Leiden, The Netherlands
[email protected]
Dr Egon Willighagen, PhD
Dept. of Bioinformatics - BiGCaT, Maastricht University
Dammestraat 5, NL-5628 NM Eindhoven, The Netherlands
[email protected]
Clemens Wittenbecher, MSc
Molecular Epidemiology, German Institute of Human Nutrition Potsdam-Rehbruecke (DIfE)
Arthur-Scheunert-Allee 114-116, 14558 Nuthetal, Germany
[email protected]
Dr Katy Wolstencroft, PhD
Leiden Institute of Advanced Computer Science, Leiden University
Niels Bohrweg 1, 2333 CA Leiden, The Netherlands
[email protected]
Dr Suzan Wopereis, PhD
Microbiology & Systems Biology, Netherlands Organisation for Applied Scientific Research (TNO)
[email protected]
Wen Wu, BSc
Laboratory of Microbiology, Wageningen University & Research Centre
Chemistry Building, building 316, Dreijenplein 10, 6703 HB Wageningen, The Netherlands
[email protected]
Prof. Ioannis Xenarios, PhD
Vital-IT/Swiss-Prot, Swiss Institute of Bioinformatics (SIB)
Quartier Sorge, bâtiment Génopode, 1015 Lausanne, Switzerland
[email protected]
Xiang Zhang, MSc
Department of pediatrics, University Medical Centre Groningen
European Research Institute for the Biology of Ageing University Medical Center Groningen Building 3226,
9700 AD Groningen, The Netherlands
[email protected]
page 26 / 88
SB@NL2014 symposium
Symposium booklet
ABSTRACTS
Abstracts
Abstracts are sorted by surname of the presenting author.
page 27 / 88
SB@NL2014 symposium
Symposium booklet
Abulikemu, Abudukelimu
Systems biology designed inhibition of mast-cell mediated tumorigenesis
Abulikemu, Abudukelimu (1); Westerhoff, Hans V. (1,2,4)
(1) Department of Molecular Cell Physiology, Institute for Molecular Cell Biology, VU University Amsterdam; (2)
Synthetic Systems Biology, Swammerdam Institute for Life Sciences, University of Amsterdam; (3) Department of
Immunopathology, Sanguin, Amsterdam; (4) Manchester Centre for Integrative Systems Biology, University of
Manchester, UK.
Cancer can be interpreted as a systems biology disease. It evolves in an interplay between oncogene
products, growth factors and their receptors, genome desintegration, signalling pathways, metabolism and
external stimuli, including inflammation. On the one hand, innate immune cells, such as resting
macrophages, mast cells and neutrophils, accumulating at the site of the tumour create a complex
“ecology” and their infiltration is associated with tumourogenesis in human cancers. On the other hand,
tumour cells secrete a wide variety of chemokines and some of these address cascades that activate
immune cells. Tumour cells for instance secrete stem cell factor (SCF), which activates mast cells, which
leads to the production of pro-angiogenic substances such as tryptase, which can enhance tumour cell COX
activity via the PAR-2 receptor. Some chemokines are beneficial to containment of the tumour cells
through the induction of acute inflammation and anti-tumour effects. Others kill fibroblasts, relieving what
is left of contact inhibition of tumour cells. As the action of mast cells may be both pro-tumorigenic and
anti-tumorigenic, we began to use systems biology to examine the potential role of mast cells in tumour
cell growth and angiogenesis. One of us had analysed mast cells in human cancer tissue arrays using antitryptase antibody and demonstrated that activated mast cells abound in cancer tissues in skin, breast,
pancreas, kidney, and lung. He had also shown that mast cells are present and de-granulated in murine B16
melanoma and human glioblastoma. In silico modelling was then designed in which the mast cells express a
specific receptor that interacts with the immunoglobulin-free light chains (FLC) after which the then
activated mast-cells secrete Tumour Necrosis Factor-alpha (TNF-α), which mediates the killing of various
cell types. Simultaneously, the activated mast cell secretes IL-4 which plays an anti-cancer role in the
tumour microenvironment through apoptotic eff.
Adriaens, Michiel
Selected Speaker
Uncovering the mechanisms modulating cardiac electrical function using systems genetics
approaches in recombinant inbred rat strains
Adriaens, Michiel (1); Moreno-Moral, Aida (2); Lodder, Elisabeth (1); Remme, Carol Ann (1); Wolswinkel,
Rianne (1); Petretto, Enrico (2); Cook, Stuart (2); Bezzina, Connie (1)
(1) Department of Experimental Cardiology, Heart Failure Research Centre, AMC, Amsterdam, Netherlands; (2)
Department of Integrative Genomics and Medicine, MRC Clinical Sciences Centre, Imperial College London, London, UK.
Although genome-wide association studies have identified many common genetic variants impacting on
susceptibility to cardiac arrhythmias and sudden cardiac death (SCD), uncovering the actual underlying
disease mechanisms remains a substantial challenge. Ideally one would combine a comprehensive
genotyping with genome-wide measurements of gene transcription, in addition to a comprehensive
phenotyping of the disease using indices of conduction and repolarization, and physical properties of the
heart. Unfortunately, such genome-wide resources for the human heart are sparse and underpowered.
Hence, the only means to paint the full picture is to complement insights derived from human studies with
page 28 / 88
SB@NL2014 symposium
Symposium booklet
systems genetics approaches in statistically powerful animal models. In this study we use 29 BXH/HXB
recombinant inbred (RI) rat strains to uncover the mechanisms modulating cardiac electrical function.
These RI strains are derived from a normotensive and spontaneously hypertensive rat strain and are one of
the most widely used models in cardiac research. Prolonged ECG indices of conduction and repolarisation
are risk factors for cardiac arrhythmias and SCD, and here we combine such indices with genetics and RNAseq transcriptomics data in the powerful BXH/HXB model. Since the complex genetic background of these
strains results in a complex dependency structure between genetic markers, a Bayesian statistical
framework is used for association analysis, yielding a posterior likelihood of association of a genetic marker
conditional on all other markers. Using this framework in combination with co-expression network analysis,
we identified multiple genes under strong cis-genetic control as well as gene networks associated with
cardiac conduction. In addition to these results, data analysis challenges and strategies for selecting
candidate genes for follow-up studies will be highlighted.
Armitage, Judith
Invited Speaker
Bacterial division: ensuring a balanced inheritance
Armitage, Judith
Department of Biochemistry, University of Oxford, UK
Most bacteria divide by septation at midcell, but this must be coordinated not only with chromosome
duplication and segregation, but also the duplication and segregation of low copy number protein
complexes and low copy number plasmids. How is duplication and segregation choreographed through the
cell cycle? Rhodobacter sphaeroides, a photosynthetic alpha proteo-bacterium has 2 chromosomes, a single
flagellum, two differently positioned chemosensory clusters and undergoes major morphological changes
when switching from aerobic to photoheterotrophic growth, but under all conditions still finds mid-cell, still
duplicates and accurately segregates the chromosomes and proteins. Combining in vivo live cell single
molecule imaging, molecular genetics and mathematical models I will discuss current ideas about how
these processes are (and are not) coordinated.
Bader, Gary
Invited Speaker
Interpreting cancer and other disease mutations in a network context
Bader, Gary D.
The Donnelly Centre for Cellular and Biomolecular Research, University of Toronto, Canada
Somatic mutations in cancer genomes include drivers that provide selective advantages to tumour cells and
passengers present due to genome instability. Discovery of pan-cancer drivers will help characterise
biological systems important in multiple cancers and lead to development of better therapies. Driver genes
are most often identified by their recurrent mutations across tumour samples. However, some mutations
are more important for protein function than others. Thus considering the location of mutations with
respect to functional protein sites can predict their mechanisms of action and improve the sensitivity of
driver gene detection. We've developed a number of computational methods and tools recently to identify
significant mutations in signalling sites, such as phosphorylations and other post-translational
page 29 / 88
SB@NL2014 symposium
Symposium booklet
modifications, as well as protein interaction sites bound by peptide recognition domains, such as SH3, WW
and PDZ domains. These methods predict how protein interaction networks are rewired by mutations and
how these may cause disease.
Balling, Rudi
Invited Speaker
Systems approaches to Parkinson's disease
Balling, Rudi
Luxembourg Centre for Systems Biomedicine, University of Luxembourg.
Parkinson´s disease (PD) is one of the major neurodegenerative diseases and primarily due to the loss of
dopaminergic neurons in the substantia nigra. The pathogenesis of PD is not well understood, however it is
thought to be multi-factorial and age-related with many genetic and environmental factors involved. In
order to identify the mechanisms involved we have initiated a pathway and network analysis of PD. In order
to capture the rapidly increasing information and inter-relationships between different factors contributing
to PD, we have developed a computationally tractable, comprehensive molecular interaction map of PD.
This map integrates pathways implicated in PD pathogenesis and captures and visualises all major
molecular pathways involved in PD pathogenesis. The PD-map can serve as a resource for further
computational analyses and as a platform for community level collaborations.
Barberis, Matteo
Selected Speaker
A Systems Biology approach to the tuning of cell cycling and chromatin dynamics
Linke, Christian (1, 2); Chasapi, Anastasia (3); González-Novo, Alberto (4); Klipp, Edda (5); Posas, Francesc
(4); Krobitsch, Sylvia (2); Xenarios, Ioannis (3); Barberis, Matteo (1, 2, 5)
(1) Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, The Netherlands; (2) Max Planck
Institute for Molecular Genetics, Berlin, Germany; (3) Swiss Institute of Bioinformatics, University of Lausanne,
Lausanne, Switzerland; (4) Departament de Ciències Experimentals i de la Salut, Universitat Pompeu Fabra, Barcelona,
Spain; (5) Institute for Biology, Humboldt University Berlin, Berlin, Germany
Networks of interacting molecules organize topology and timing of biological functions. The cell cycle is a
time-dependent process, which relies on events that are kept separate in time, i.e. DNA replication, cell
division. Cyclin-dependent kinases (Cdks) mediate a faithful timing of these events, by binding to pools of
cyclins in successive waves of expression. This coordination is such that waves of cyclin/Cdk activity occur
sequentially at different times throughout cell cycle progression. However, how the precise timing of cyclin
waves is managed, is not understood. Here we present our findings addressing the molecular basis of
mitotic (Clb) cyclin oscillations in budding yeast, with the focus on regulatory motifs that may generate
timely waves. By integrating computational techniques (kinetic and logical modelling) with detailed
experimental investigations (protein-protein interaction and biochemical assays, chromatin
immunoprecipitation, time-resolved dynamics and live cell imaging) we investigate the modulators of the
network regulating waves of cyclin/Cdk activity in the budding yeast Saccharomyces cerevisiae. Coupled
activities of B-type cyclins bound to Cdk1 and transcription are investigated, with a major focus on the
molecular basis of cyclin oscillations. A definite transcriptional regulation synchronizing the temporal
appearance of phase-specific Clb cyclins is unravelled. Furthermore, regulatory motifs are pinned down,
that may help to generate timely Clb waves. In summary, a multidisciplinary, systems biology approach
page 30 / 88
SB@NL2014 symposium
Symposium booklet
integrating experimental analyses into appropriate computational frameworks can provide a powerful tool
to pinpoint regulatory modules to fine tune the precise cell cycle timing, highlighting design principles in
cell division.
Barneh, Farnaz
Data completion: an important determinant for successful integration of drug-target and disease
networks
Barneh, Farnaz (1); Jafari, Mohieddin (2)
(1) Faculty of Paramedical Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran; (2) Protein Chemistry
& Proteomics Unit, Biotechnology Research Centre, Pasture Institute of Iran, Tehran, Iran.
Motivation
Dynamic nature of pharmaceutical industry encourages continuous assessment of data completion for
effective integration of drug-target and disease networks. We aim to analyse evolutionary changes in drugtarget network topology following information updates in DrugBank database.
Methods
List of approved drugs and protein targets in DrugBank1.0 and 4.0 were downloaded from DrugBank
database. Drug-target network was constructed and analysed using Gephi. Drug similarity network
projection was also analysed. Cellular components of targets in two DrugBank versions were evaluated by
BiNGO.
Results
Number of connected components was 209 and 108 in DrugBank1.0 and 4.0. Modularity index increased to
0.907 and 0.81 with 14 versus 26 modules in giant components respectively. Distribution of drug categories
in giant component also changed. In drug-network projection, many drugs moved into giant component
following newly added data in DrugBank 4.0. Cellular membrane was still main target site for drugs.
Mitochondrion and nucleus were also seen in DrugBank 4.0 target sites.
Conclusion
As knowledge about number of protein targets per each drug increases, central topological features such as
‘hubs’, ‘bridges’, number and size of ‘connected components’ change in the whole network; all of which
serve as critical determinants of network-based drug discovery and integration into disease networks.
Benis, Nirupama
Network analysis of temporal functionalities of the gut induced by perturbations in new-born
piglets.
Benis, Nirupama (1); Schokker, Dirkjan (2); Suarez-Diez, Maria (3); Martins dos Santos, Vitor (3,4); Smidt,
Hauke (5); Smits, Mari (1,2,6)
1. Host Microbe Interactomics, Wageningen University, 2. Wageningen UR Livestock Research, 3. Systems and
Synthetic biology, Wageningen University, 4. LifeGlimmer GmbH, Berlin, Germany, 5. Laboratory of Microbiology,
Wageningen University, 6. Central Veterinary Institute of Wageningen UR.
Background
page 31 / 88
SB@NL2014 symposium
Symposium booklet
Evidence is accumulating that perturbation of early life microbial colonization of the gut induces longlasting adverse health effects in individuals. Understanding the mechanisms behind these effects will
facilitate modulation of intestinal health. The objective of this study was to identify biological processes
involved in these long lasting effects and the (molecular) factors that regulate them. We used an antibiotic
and an antibiotic in combination with stress on piglets as an early life perturbation. Then we used host gene
expression data from the gut (jejunum) tissue and community-scale analysis of gut microbiota from the
same location of the gut on three different time points to gauge the reaction to the perturbation. We
analysed the data in two dimensions, treatment and time, with quadratic regression analysis. Then we
applied network based, data integration approaches to find correlations between host gene expression and
the resident microbial species.
Results
We observed significant long-lasting differences in jejunal gene expression patterns resulting from the early
life perturbations. In addition, we were able to identify potential key gene regulators (hubs) for these longlasting effects. Data integration also showed that there are a handful of bacterial groups that were
associated with temporal changes in gene expression.
Conclusion
The applied approach allowed us to unearth biological processes involved in long lasting effects in the gut
due to the early life perturbation of microbial colonization. The observed data are consistent with the
hypothesis that these long lasting effects are most probably due to differences in the programming of the
gut immune system as induced by the temporary early life changes in the composition and / diversity of
microbiota in the gut.
Boas, Sonja
Selected Speaker
Tip cell overtaking arises spontaneously in computational models of angiogenic sprouting
Boas, Sonja E. M. (1); Merks, Roeland M. H. (1,2)
(1) Life Sciences group, CWI, Amsterdam; (2) Mathematical Institute, Leiden University, The Netherlands.
The traditional view of angiogenesis, the formation of new blood vessels from existing ones, is that
endothelial cells differentiate into tip and stalk cells, after which one tip cell leads the sprout. More
recently, it was shown that cells compete for the tip position during angiogenic sprouting [1,2]: a
phenomenon named tip cell overtaking. Experimental and computational results [1,3] suggest that tip cell
overtaking is regulated by an interplay of VEGF and Delta-Notch signalling. In contrast, another
experimental group [2] showed that tip cell overtaking occurs independently of VEGF and Delta-Notch
signalling, putting the signalling-mediated control of tip cell overtaking into question. To address this
debate, we studied tip cell overtaking in two alternative, minimal computational models, in which
angiogenic sprouting is driven by generic cell behaviours: contact-inhibited chemotaxis or elongated cell
shape and chemotaxis, respectively. In line with experimental observations, simulation results show that
cells migrate forward and backward within sprouts (Figure 1), also called cell-mixing. We found that tip cell
overtaking occurs spontaneously in both models simply as a result of cell-mixing, independent of DeltaNotch or VEGF signalling. In addition, the contact-inhibited chemotaxis model supports the experimental
observations [1] that Delta-Notch and VEGF signalling can influence the potency of cells to occupy the tip
position when cells in the sprout differ in their sensitivity to VEGF. In a previous computational study [3]
was suggested that cells with a relative high VEGF sensitivity increase their potential to occupy the tip by
VEGFR2-mediated reduction of cell-cell adhesion. In addition to this mechanism, our model suggests a
mechanism in which VEGF sensitive cells become less sensitive to an auto-secreted chemo-attractant that
drives sprouting.
page 32 / 88
SB@NL2014 symposium
Symposium booklet
Boer, Wim P.H. de
Program Curfit2D: two-dimensional alignment of GC-GC chromatograms
De Boer, Wim P.H. (1); Lankelma, Jan (1)
(1) Other
The computer program Curfit2D [1] for the two-dimensional alignment of GC-GC or LC-LC chromatograms is
now available [2]. The program can also be used to align LC-MS chromatograms by running the program in
“one-dimensional” mode: the mass spectra are aligned alongside the alignment of the LC-axis, thereby
maintaining the distinction between different peaks and retaining the discrete distribution of the MS
values. The Curfit2D algorithm is based on the idea to model shifts with a two-dimensional “warp function”
such that the sample chromatogram, its shifts corrected with the warp function, is adjusted to the
reference chromatogram by minimizing the squared intensity difference. The benefits of the present
alignment procedure are: (a) alignment of highly non-linear shifts; (b) easy handling of overlapping
peaks; (c) robust against missing peaks and concentration variation of compounds. Missing peaks can be
tolerated if their absence has little numerical effect on the warp function computation and if they occur
between existing peaks. Program Curfit2D is fitted with two pre-processing algorithms for baselinecorrection and data-smoothing, respectively. Both algorithms are based on a moving window of minimal
size, 5 x 5 in practice. The window sizes are specified in terms of data measuring points and must be oddnumbered in order to preserve the symmetry of peaks. The data-smoothing approach presented here
lowers the height of a peak and broadens its base, but the peak’s volume is hardly affected. Tests
performed on single-peak model chromatograms support this observation. The poster presents heat maps
of a diesel oil chromatogram pair before and after alignment. All runs were done on a 64-bit workstation
under Windows 7.
References
[1] Wim P.H. de Boer, Jan Lankelma, “Two-dimensional semi-parametric alignment of chromatograms”,
Journal of Chromatography A, 1345 (2014) 193-199
[2] http://www.falw.vu/~microb/mcf/research/GS4/index_Curfit2D.html.
Bohler, Anwesha
Combining measured or modelled metabolite flux data with omics data on biological pathways
page 33 / 88
SB@NL2014 symposium
Symposium booklet
Bohler, Anwesha Bohler (1,2); Pamu Sriharsha (3); Guo, Ruizhou(4); Van Iersel, Martijn P. (5); Evelo, Chris T.
(1,2,6)
1. Department of Bioinformatics - BiGCaT, NUTRIM, Maastricht University, The Netherlands. 2. Netherlands
Consortium for Systems Biology, The Netherlands. 3. College of Science, Illinois Institute of Technology, USA 4.
Department of Biomedical Engineering - CBio Group, Eindhoven University of Technology, The Netherlands. 5. General
Bioinformatics, United Kingdom. 6. Maastricht Centre for Systems Biology, Maastricht University, The Netherlands.
Biological pathways are used for integrative analysis of transcriptomic, epigenomic, proteomic, and
metabolomic measurements. Using our open-source pathway analysis platform, PathVisio, we have
bridged pathway analysis to quantitative approaches used for metabolic network modelling, such as flux
balance analysis and dynamic simulation. Our focus is on the visualisation of the modelling results in the
same context as the experimental data visualisation. This is critical for understanding how simulated
models correlate with experimental measurements and helps gain more insight into the mechanism of the
biological processes. PathVisio enables researchers to draw pathways as diagrams and annotate the genes,
proteins, and metabolites in the pathway diagram with database identifiers. This enables data to be
mapped and visualised on the pathway diagram. An identifier mapping database is used by PathVisio to
map the data uploaded by the user to the pathway irrespective of the identifier system used. We have
extended PathVisio to enable annotation of interactions and created an interaction identifier mapping
database based on the EBI-Rhea database. Furthermore, two plugins for PathVisio, PathSBML and IntViz,
were made to convert quantitative models from BioModels stored as SBML into the format used by
PathVisio and to visualise flux data on interactions. Pathways from WikiPathways, an open platform for
curating, sharing and publishing biological pathways, can be used in PathVisio. Therefore, these new
developments allow data to be visualised on interactions alongside gene products on all WikiPathways
pathways. This allows researchers to show modelling results on pathways along with genomics data and
visualise the models and results themselves in combination with omics data. This kind of integrated
visualisation makes modelling results more accessible and interpretable to biologists who wish to learn
from them and also facilitates new ways to explore the processes involved.
page 34 / 88
SB@NL2014 symposium
Symposium booklet
Bonder, Marc Jan
The influence of a short-term gluten free diet on the human microbiome
Bonder, MJ (1)*, Cai, X (3)*, Tigchelaar, EF (1,2)*, Trynka, G (1,4), Cenit, MC (1), Hrdlickova, B (1), Zhong, H
(3), Gevers, D (4), Wijmenga, C (1,2), Wang, Y (3)#, Zhernakova, A (1,2)#
(1) University of Groningen, University Medical Centre Groningen, Department of Genetics, Groningen, the
Netherlands; (2) Top Institute Food and Nutrition, Wageningen, the Netherlands; (3)
BGI-Shenzhen, 518083, China;
(4)
Broad Institute of MIT and Harvard, Cambridge, MA, US.
A gluten free diet (GFD) is the most common diet worldwide. It is not only an effective treatment for celiac
disease, but also commonly followed by individuals with intestinal discomfort. There is an important link
between diet and microbiome, although how GFD affects the human microbiome is largely unknown.
Therefore, we wanted to get more insight into the influence of a short-term GFD on the human
microbiome. We studied changes in the gut microbiome in 21 healthy volunteers followed a GFD for four
weeks. Stool samples were collected before the start of the diet, then at weekly intervals during the GFD,
and after a wash-out period of five weeks again at weekly intervals for four weeks on habitual diet (HD) (1
HD + 4 GFD + 4 HD). Microbiome profiles were determined using 16s rRNA sequencing and subsequently
processed for taxonomic and imputed functional composition. In addition, the levels of six gut health
related biomarkers in plasma or feces were assessed at the same time points as the stool sample collection.
The short-term GFD intervention in healthy subjects had a limited overall impact on the microbiome,
maintaining the interpersonal diversity observed prior to the start of diet. However, a number of taxonspecific differences were noticed; the most striking shift was seen for the family Veillonellaceae (class
Clostridia) that significantly reduced during the intervention (p=2.8e−05, q=0.003), Figure 1. Additionally,
seven other taxa were significantly changed, several of them are known to play a role in starch metabolism.
Differences in pathway activities in relation to diet were more pronounced; twenty-one predicted pathway
activity scores showed significant association to the change in diet, with four of the top five pathways were
related to metabolism. In conclusion, we observed significant changes in microbiome composition and
microbiota pathways activities associated to a short-term GFD.
page 35 / 88
SB@NL2014 symposium
Symposium booklet
Breda, Simone van
Epigenetic mechanisms underlying arsenic-associated lung carcinogenesis
Van Breda, Simone (1); Claessen, Sandra (1); Lo, Ken (2); van Herwijnen, Marcel (1); Brauers, Karen (1);
Lisanti, Sofia (3); Theunissen, Daniël (1); Jennen, Danyel (1); Gaj, Stan (1); de Kok, Theo (1); and, Kleinjans,
Jos (1).
(1) Department of Toxicogenomics, GROW School for Oncology and Developmental Biology, Maastricht University
Medical Centre+, Maastricht, P.O. Box 616, 6200 MD, the Netherlands; (2) Roche Applied Science, Madison, WI 53719,
United States of America; (3) Human Nutrition Research Centre, Institute for Ageing & Health, Newcastle University,
Newcastle upon Tyne, NE4 5PL, UK.
Arsenic is an established human carcinogen, but the mechanisms through which it contributes to for
instance lung cancer development are still unclear. As arsenic is methylated during its metabolism, it may
interfere with the DNA methylation process, and is therefore considered to be an epigenetic carcinogen. In
the present study, we hypothesize that arsenic is able to induce DNA methylation changes, which lead to
changes in specific gene expression, in pathways associated with lung cancer promotion and progression.
A549 human adenocarcinoma lung cells were exposed to a low (0.08 µM), intermediate (0.4 µM) and high
(2 µM) concentration of sodium arsenite for 1, 2 and 8 weeks. DNA was isolated for whole genome DNA
methylation analyses using NimbleGen 2.1M deluxe promoter arrays. In addition, RNA was isolated for
whole genome transcriptomic analysis using Affymetrix microarrays. Using a systems biology approach,
extensive software applications were used in order to investigate methylation differences between
different conditions and additional integrative analyses with transcriptomics data. Arsenic modulated DNA
methylation and expression levels of hundreds of genes in a dose- and time-dependent manner. By
performing cross-omics analyses, in which whole genome DNA methylation and gene expression data with
possibly involved transcription factors were combined, a large molecular interaction network was created
based on transcription factor-target gene pairs, consisting of 216 genes. A tumour protein p53 (TP53) sub
network was identified, showing the interactions of TP53 with other genes affected by arsenic.
Furthermore, multiple other new genes were discovered showing altered DNA methylation and gene
expression. In particular, arsenic modulated genes which function as transcription factor, thereby affecting
target genes which are known to play a role in lung cancer promotion and progression.
Cavill, Rachel
Four classes of methods for integrating transcriptomic and metabolomic data
Cavill, Rachel; Briede, Jacco Jan; Kleinjans, Jos
Department of Toxicogenomics, Maastricht University.
Transcriptomic and metabolomic data give complementary information about biological systems. However
the integration is complex and often confusing. Often these data are investigated by performing separate
analyses, and comparing results. But this tends to highlight differences rather than similarities. Integrative
analysis methods combat this, four classes of these are described below. In concatenative analysis, one
combines the datasets into a single table and performs the usual analysis. However, this often fails, as the
datasets will have different underlying structures and the data and noise is drawn from different
page 36 / 88
SB@NL2014 symposium
Symposium booklet
distributions. For instance, a metabolomics and transcriptomics dataset concatenated and then clustered
will split into two clusters, the metabolites and the transcripts. Correlations hold when looking across
similar conditions, but correlations between genes and metabolites can reverse in direction when
conditions change significantly (e.g. Aerobic vs anaerobic metabolism) so care must be taken. When time
series are present, changes are not simultaneous, so techniques such as dynamic time warping (e.g.
DTW4omics in R) are used to compensate. Multivariate models are powerful, but are not easy to build and
interpret by non-experts. One technique O2PLS is symmetric. So instead of modelling metabolomics from
transcriptomics or vice versa, the models and predictions are made in both directions. The output gives the
% variance of each dataset which is explained by the other. This is useful for highlighting where similarities
between the datasets. Pathway analysis is useful as the output is biological rather than mathematical and it
is intuitive for biologists to use and interpret. IMPaLA (impala.molgen.mpg.de) is an example that performs
either over-representation or enrichment analysis on metabolomics and transcriptomics data jointly. This
means that more pathways can be found by using evidence of change from both datasets.
Chasapi, Anastasia
Boolean modelling of S. pombe septation initiation network
Chasapi, Anastasia (1); Wachowicz, Paulina (2); Niknejad, Anne (1); Simanis, Viesturs (2); Xenarios, Ioannis
(1,3)
(1) Vital-IT Group, Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland; (2) Cell cycle control laboratory, Ecole
Polytechnique Fédérale de Lausanne (EPFL), SV-ISREC, Lausanne, Switzerland; (3) Swiss-Prot Group, Swiss Institute of
Bioinformatics (SIB), Geneva, Switzerland
Schizosaccharomyces pombe, commonly called fission yeast, has long been used as a model organism in
cell cycle studies, due to the similarities between its cell cycle and the mammalian cell cycle. The Septation
Initiation Network (SIN), a protein kinase signalling network using the spindle pole body as scaffold,
controls cytokinesis in fission yeast. Our goal is to describe the qualitative behaviour of the system and
potentially conditional dependencies between different mutants and their ability to form a functional SIN.
Towards this end, we decided to adopt a qualitative modelling approach. Prior to this work, models of the
SIN in fission yeast had been published; a quantitative model by Csikasz-Nagy et al. (2007) and a minimal
qualitative model by Bajpai et al. (2013). Both models reveal important regulatory aspects of the SIN but
they include a minimal set of SIN regulators. We report the construction of an extended, Boolean model of
the SIN, including most SIN regulators as experimentally testable entities. The model uses the varying cell
cycle CDK levels as control nodes for the simulation of SIN related events. All model simulations were
performed with BoolSim, a software tool for synchronous and asynchronous Boolean modelling of gene
regulatory networks, based on reduced ordinary decision diagrams (ROBDDs) (Garg et al. 2008). The model
was optimized using a training set of in silico single knock-out experiments of known phenotypic effects,
and it was able to predict the in silico experimental test set, as well as provide useful insights for the SIN
regulation.
Coort, Susan
Selected Speaker
A systems biology approach to study transcriptomics data of the diabetic liver
Kutmon, Martina (1,2); Chris T Evelo (1); Susan L Coort (1)
page 37 / 88
SB@NL2014 symposium
Symposium booklet
1. Department of Bioinformatics - BiGCaT, NUTRIM School for Nutrition, Toxicology and Metabolism, Maastricht
University, the Netherlands; 2. Maastricht Centre for Systems Biology (MaCSBio), Maastricht University, the
Netherlands.
Nowadays a broad collection of transcriptomics data is publicly available. Data analysis methods often aim
at deciphering the influence of gene expression at the process level. Biological pathway diagrams depict
known processes and capture the interactions of gene products and metabolites, information that is
essential for the interpretation of transcriptomics data. The present study describes a comprehensive
systems biology approach that integrates differential gene expression in the human diabetic liver with
pathway information by building a network of interconnected pathways. A dataset comparing the liver
transcriptome from lean, healthy vs. obese, insulin-resistant subjects was selected. Pathway analysis
revealed 7 significantly altered pathways. These pathways were then merged into one combined network
with 408 gene products, 38 metabolites and 5 pathway nodes. Further analysis highlighted 17 nodes
present in multiple pathways, and revealed the connections between different pathways in the network.
The integration of transcription factor-gene interactions from the ENCODE project identified new links
between the pathways on a regulatory level, see figure. In addition, the extension of the network with
known drug-target interactions from DrugBank allowed to study drug actions and helped with the
identification of other drugs that target proteins up- or downstream. The described systems biology
approach uses state-of-the-art pathway and network analysis methods to study the rewiring of the diabetic
liver. The integration of experimental data and knowledge on disease-affected biological pathways,
including regulatory elements like transcription factors or drugs, leads to improved insights and a clearer
illustration of the overall process. It also provides a resource for building new hypotheses for further
follow-up studies. The approach is highly generic and can be applied in different research fields. Accepted
for publication in BMC Genomics.
page 38 / 88
SB@NL2014 symposium
Davids, Mark
Symposium booklet
Poster Flash Presenter
Functional profiling of unfamiliar microbial ecosystems
Davids, Mark (1,5); Hugenholtz, Floor (2,5); Martins dos Santos, Vitor (1,5); Smidt, Hauke (2,5);
Kleerebezem, Michiel (3,4,5); Schaap, Peter (1,5).
(1) Laboratory of Systems and Synthetic Biology, Wageningen University, Dreijenplein 10, 6703 HB Wageningen, The
Netherlands; (2) Laboratory of Microbiology, Wageningen University, Dreijenplein 10, 6703 HB Wageningen, The
Netherlands; (3) Host-Microbe Interactomics Group, Wageningen University, P.O. box 338, 6700 AH, Wageningen, The
Netherlands; (4) NIZO Food Research B.V, P.O. Box 20, 6710 BA, Ede, The Netherlands Research B.V, P.O. Box 20, 6710
BA, Ede, The Netherlands; (5) Netherlands Consortium for Systems Biology; TI Food and Nutrition, P.O. Box 557, 6700
AN Wageningen, The Netherlands
Analysis of complex metatranscriptome datasets of natural microbial communities poses a considerable
bioinformatics challenge since they are essentially non-restricted with a varying number of participating
strains and species. Metatranscriptome studies using RNA-Seq usually apply a read-to-gene mapping
strategy to determine the taxonomic origin and function encoded within the sequenced transcripts. When
dealing with well-studied environments that have good reference sequences overall good results are
obtained. However with increasing evolutionary distance between the reference sequences and the
sampled microbial community the overall alignment rates decrease resulting in a biased view of microbial
composition and overall activity. To by-pass this, peptide based alignments are being used. However this
method is computationally much more intensive than nucleotide based alignments and due to the use of
short read sequences translated peptide-based alignments may result in many false positive assignments.
Assembly of short RNA-Seq reads leads to an increase in information content. This would result in a more
robust functional and taxonomic assignment at an increased evolutionary distance and a significant
reduction in computational time required. In complex natural communities assembly may result in short
contigs and chimeric constructs and low abundant transcripts may not be detected. Here we report on the
development of a de novo assembly workflow for analysis of metatranscriptome data that was used for the
analysis of presently uncharacterized mouse cecal communities providing insight in global, and family
specific activities. We monitored precision and recall of the workflow at sequence, functional and
taxonomic level using control RNA-Seq metatranscriptome datasets. The results indicate that with this de
novo metatranscriptome assembly pipeline RNA-Seq reads from unfamiliar ecosystems can be assigned to a
protein function and a taxonomic rank with high precision.
Deelen, Patrick
Selected Speaker
Calling genotypes from public RNA-sequencing data enables identification of genetic variants
that affect gene-expression levels
Deelen, Patrick (1, 2, *); Zhernakova, Daria V. (1, *); de Haan, Mark (1, 2); van der Sijde, Marijke (1); Bonder,
Marc Jan (1); Karjalainen, Juha (1); van der Velde , K. Joeri (1, 2); Abbott, Kristin M. (1); Fu, Jingyuan (1);
Wijmenga, Cisca (1); Sinke, Richard J. (1); Swertz, Morris A. (1, 2, #); Franke, Lude (1, #)
(1) University of Groningen, University Medical Centre Groningen, Department of Genetics, Groningen, the
Netherlands; (2) University of Groningen, University Medical Centre Groningen, Genomics Coordination Centre,
Groningen, The Netherlands; (*,#) These authors contributed equally to this work.
Introduction
page 39 / 88
SB@NL2014 symposium
Symposium booklet
RNA-sequencing (RNA-seq) is a powerful technique for the identification of genetic variants that affect gene
expression levels, either through expression quantitative trait locus (eQTL) mapping or through allele
specific expression (ASE) analysis. Given increasing numbers of RNA-seq samples in the public domain, we
here studied to what extent eQTLs and ASE effects can be identified when using public RNA-seq data while
deriving the genotypes from the RNA sequencing reads itself.
Results
4,978 public human RNA-seq runs, representing many different tissues and cell-types, passed quality
control (Figure 1). Even though this data originated from many different laboratories, samples reflecting
the same cell-type clustered together, suggesting that technical biases due to different sequencing
protocols are limited. We derived genotypes from the RNA-sequencing reads and imputed non-coding
variants. In a joint analysis on 1,262 samples combined, we identified cis-eQTLs effects for 8,034 unique
genes (at a false discovery rate ≤ 0.05). eQTL mapping on individual tissues revealed that a limited number
of samples already suffice to identify tissue-specific eQTLs for known disease-associated genetic variants.
Additionally, we observed strong ASE effects for 34 rare pathogenic variants, corroborating previously
observed effects on the corresponding protein levels.
Conclusion
By deriving and imputing genotypes from RNA-seq data, it is possible to identify both eQTLs and ASE
effects. Given the exponential growth of the number of publicly available RNA-seq samples, we expect this
approach will become especially relevant for studying the effects of tissue specific and rare pathogenic
genetic variants to aid clinical interpretation of exome and genome sequencing.
Figure 1. PCA on the gene-expression levels reveal clear discrimination between primary tissues, cell line
samples and hematopoietic sample.
Dijk, Aalt-Jan van
Selected Speaker
Predicting flowering time via data-integration using a combined statistical-mathematical model
Van Dijk, Aalt D.J. (1); Molenaar, Jaap (1)
(1) Biometris, Mathematical and Statistical Methods, Wageningen University
page 40 / 88
SB@NL2014 symposium
Symposium booklet
Flowering at the right moment is crucial for the reproductive success of plants. Hence, plants integrate
various environmental cues with endogenous signals in order to flower under optimal conditions. We
recently published a first dynamic model for the network involved in integration of those signals [1]. This
model describes how a set of eight transcription factors (TFs) regulate each other’s expression. To fit the
model parameters we exploited time course expression levels, data describing regulatory interactions, and
phenotypic data (flowering time of various mutants). The model was successful in predicting the effect of
perturbations in expression levels of these TFs on flowering time. In the present project we connect the
integration network model to a network model for the various upstream genes that are involved in
receiving environmental and endogenous signals. Data available for the latter does not consist of
expression times series, but of large amounts of gene expression data, measured under various conditions,
typically of static nature. We thus face a typical data-integration problem. Our approach is to use for the
upstream network a statistical model, while the integration network is formulated in terms of ODEs. We
demonstrate how to connect a Bayesian Network (BN) model describing connections between hundreds of
upstream genes, with a detailed dynamic ODE model which receives input from the BN. By connecting a
large statistical model to a detailed smaller mathematical model we are able to integrate very different
type of data and to tackle the complexity of the system. The BN predicts the effect of perturbations of any
of the upstream genes on genes included in the integration network ODE model. Using the signal from the
BN as input, the ODE model predicts the resulting change in flowering time. We demonstrate the predictive
power of this approach by applying the model to mutant data. [1] Leal Valentim et al., PLoS ONE, in press.
Driel, Roel van
Invited Speaker
Towards understanding biological systems: exciting science and societal benefits
Van Driel, Roel (1,2)
(1) Synthetic Systems Biology and Nuclear Organisation Group (NOG), Swammerdam Institute for Life Sciences (SILS),
Faculty of Science (FNWI), University of Amsterdam; (2) Infrastructure Systems Biology Europe (ISBE).
Changes are needed in the life sciences and its research traditions to realise the potential of systems
biology. The move from compounds biology to the biology of large complex systems calls for not merely a
change of perspective, but also a change of the institutional settings at which biology research is
conducted. The scientific world needs to attune to the development of larger well-coordinated research
programs, which entails better coordination between research groups. One of the key issues is that the
field needs to commit to a standardisation of experimentation and data, the lack of which hinders
compatibility and synergy between research partners. It also means that traditional ideas about career
paths, from PhD to postdoc to assistant, associate and full prof need to be reassessed. Young researchers
need to be prepared for novel types of career development. Another key issue is that systematically
tackling - and eventually understanding - large biological systems (cells, tissues, organs, and complete
organisms) will have a major impact on society, including industry and bio-economy, the same society that
is funding the life sciences community. Therefore, young scientists should be prepared for a double role
between science, industry and society. This session will discuss the consequences of these issues for life
sciences in general and the systems biology community in particular.
Eunen, Karen van
page 41 / 88
SB@NL2014 symposium
Symposium booklet
Inborn errors in fatty-acid metabolism make the beta-oxidation hypersensitive to substrate
overload
Van Eunen, Karen (1); Touw, Nienke (1); Gerding, Albert (1); Bleeker, Aycha (1); Heiner, Rebecca M (2);
Derks, Terry G.J. (3); Reijngoud, Dirk-Jan (1,2); Groen, Albert K. (1); Bakker, Barbara M (1)
1 Department of Pediatrics, Centre for Liver, Digestive and Metabolic Diseases, University of Groningen, University
Medical Centre Groningen, The Netherlands; 2 Department of Laboratory Medicine, Centre for Liver, Digestive and
Metabolic Diseases, University of Groningen, University Medical Centre Groningen, The Netherlands; 3 Section of
Metabolic Diseases, Beatrix Children’s Hospital, University of Groningen, University Medical Centre Groningen, The
Netherlands
Recently, we showed by dynamic modelling that its pathway structure makes the fatty-acid beta-oxidation
intrinsically vulnerable to substrate overload: at a high influx of palmitoyl-CoA into the pathway the flux
dropped and intermediate CoA-esters accumulated extremely (Van Eunen et al., 2013 PloS Comp Biol).
Although it is unlikely that this happens in a healthy cell, we show here that inborn enzyme deficiencies
might aggravate the risk. We used the previously constructed dynamic model to study the inborn enzyme
deficiency medium-chain acyl-CoA dehydrogenase deficiency (MCADD). MCADD is the most common
inborn error in the mitochondrial fatty fatty-acid. The MCAD enzyme is one of the enzymes responsible for
the first step in the fatty-acid beta-oxidation and is specific for the medium-chain acyl CoAs (C4-C12). The
availability of an MCAD-knockout mouse gave us the opportunity to study the relation between metabolic
profile and the sensitivity towards substrate overload in detail. Oxygen consumption rates and metabolite
concentrations were measured over time in isolated liver mitochondria from MCAD-knockout or wild type
mice upon addition of palmitoyl-CoA (C16) or octanoyl-CoA (C8) as substrate. Besides, mitochondrial
protein levels and activities of beta-oxidation enzymes were measured. The lack of alterations between the
wild type and MCAD-KO mice at the level of protein and enzyme activity suggests that the loss of MCAD
was not compensated. Moreover, the dynamic experiments show the robustness of the pathway against
the enzyme deficiency, since beta-oxidation maintained a substantial flux in the MCAD-KO. Nevertheless,
after fitting of model to the experimental data, the model predicted that the loss of MCAD caused an
increased sensitivity for substrate overload. We conclude that the pathway structure of the beta-oxidation
in which substrates compete for enzymes, may be at the basis of the disease phenotypes associated with
enzyme deficiencies.
Evelo, Chris
Invited Speaker
Climbing the path to math. Step by step through a data maze.
Evelo, Chris
BiGCaT group, Department of Bioinformatics, Faculty of Health, Medicine and Life Sciences, Maastricht University
(UM).
Starting from massively parallel omics data, for instance full transcriptome data, it is quite a long way
before you arrive at a mathematical model that describes and predicts what happens in a living being under
the conditions studied. Analysis of such data suffers from oversampling. After all if you measure the
expression of 20,000 genes some of them are bound to change by chance. False discovery rate corrections
will lead to some loss of real findings. The sheer number of biomolecular entities measured in genome wide
experiments also makes interpretation hard. 20,000 Terms correspond to about a tenth of the content of a
modern dictionary, and you really do not want to read all that.
page 42 / 88
SB@NL2014 symposium
Symposium booklet
One approach to lessen both these problems is to analyse such large datasets in the context of what we
already know. Biologists often use pathways to explain and visualize what is known about biological
processes. Pathways are used for instance in papers, presentations, and textbooks. Some well-known web
resources for computer readable versions of biological pathways are KEGG, Reactome, and our own
WikiPathways. These pathways are biological data models but connection of experimental data to such
pathway models needs mapping. Typically both the entities in experimental data that we want to analyse
and the “same” entities in pathways have some kind of associated database identifiers. But these
identifiers are often from different databases and sometimes not even describing the exact same entity.
The use of a microarray probe set identifier reporting for mRNA expression while the pathway contains
protein identifiers is a typical example. Such mappings are in fact needed for almost every data integration
step. BridgeDb is used for identifier mapping in our pathway tool PathVisio and in WikiPathways. It can also
be used in tools like Cytoscape, Bioconductor/R, Open PHACTS or as a web service. Using this we can
perform pathway enrichment statistics and pathway data visualization for any kind of gene product or for
metabolomics data.
Two new developments allow us to combine these pathway approaches with mathematical modelling.
With the SBML converter mathematical models can be converted into WikiPathways pathway format. With
a flux/interaction visualization plugin for PathVisio, modelling results can be visualized on these pathways.
This, of course, also needed new identifier mapping databases for biological reactions. One big advantage is
that visual representations of models can now be updated automatically with the model itself, which makes
a critical evaluation of the models much easier. The other advantage is of course that experimental data
can be combined with modelling results and visualized on the same pathway model representation.
Modelling of effects of measured transcriptomics changes on predicted enzyme kinetics could thus for
instance be combined with actual and predicted metabolite concentrations all in one pathway.
Still, neither pathway nor simulation models can contain all the known regulatory interactions for the
processes they describe. That is why we will often want to use network biology approaches to extend
pathways dynamically with things like directly interacting proteins, transcription factors, regulatory RNAs,
or drugs. With the WikiPathways plugin for Cytoscape we can analyse pathways, including the SBML
converted models, as networks and thus use network approaches to evaluate which regulatory processes
might affect the models studied. Effectively this connects another big data world to data modelling and
simulations.
All these approaches will become even more powerful if we learn to integrate data about genetic variations
better. Once we learn to map genetic changes to the specific interactions that they affect in a biological
network we learn to understand how different variations affecting the same process can lead to epistatic
interactions or how specific variations in the same gene can lead to “edgetics” because they affect different
edges in the network underlying the pathway. To do this we have to integrate tools that predict effects of
genetic variations on specific interactions to “move” these gene related variations to the edges that they
affect. Combining all the aspects described above we could also model the effects of such changes in
interactions caused by genetic variations and use that for predictions.
Of course the things described in the last paragraphs are still futuristic in part. But many of the needed
steps can already be made. As usual the path is easier to find if you walk it backwards. Converting a
mathematical model to a pathway allows evaluation of different types of experimental data and modelling
results in the same context. The maze is completed by adding regulatory information and genetic data
affecting interactions. Both these types of added information can then be evaluated through changes made
in the mathematical model or its parameters. We need to learn to walk all the steps that form this path,
instead of focussing on single steps. In this way biologists may still learn how to fix a radio.
Félix Garza, Zandra C.
Blue light phototherapy for Psoriasis from a systems biology perspective
page 43 / 88
SB@NL2014 symposium
Symposium booklet
Félix Garza, Zandra C. (1,2); Liebmann, Joerg (2); Hilbers, Peter A.J. (1); van Riel, Natal A.W. (1)
(1) Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven; (2) Light & Health at Philips
Lighting BV, Eindhoven.
This work analyses the effect of UV-free blue light (BL) irradiation of the skin using mathematical modelling.
Prior research has shown that blue light reduces the proliferation of keratinocytes by inducing their
differentiation, and causes apoptosis of lymphocytes. The effects of blue light on these cells make it an
attractive phototherapy alternative for inflammatory skin conditions, such as psoriasis. Nevertheless, the
exact process by which BL affects these cells is not fully understood. A modelling approach may give further
insight to understanding how BL irradiation of psoriatic skin leads to the control of the disease. However,
no mathematical model is available describing this phenomenon. Two deterministic models were therefore
made to describe the epidermal kinetics and interaction between keratinocytes and lymphocytes under the
effect of BL irradiation; focusing mainly on the case of psoriasis. We employed a systems biology approach
to characterize the effect of BL irradiation of the skin. Since in phototherapy parameters such as fluence
and power have a strong impact on the outcome, a parameter sensitivity analysis (PSA) was performed to
estimate a range of fluence and power at which BL phototherapy could be successful. The models results
suggest that the management of psoriasis is achieved by inducing symmetric differentiation of the
keratinocytes in the epidermal proliferative compartment. It is observed that BL irradiation of psoriatic skin
decreases the density of keratinocytes and transiently increases the density of lymphocytes, leading to the
regulation of the interaction between these two cell types. The PSA of the models predicts that the higher
the peak power the better the outcome of the BL phototherapy with a dose of 90J/cm2 per day. This
systems biology approach provides additional insight into the use of BL phototherapy for inflammatory skin
disorders.
Franke, Lude
Invited Speaker
Identification of downstream effects for many genetic risk factors by reanalysing gene
expression data
Franke, Lude (1)
(1) Department of Genetics, University Medical Centre Groningen (UMCG).
Text TBA
Greef, Jan van der
Invited Speaker
East is East and West is West: and never the twain shall meet?
Jan van der Greef
Sino-Dutch centre for Preventive and Personalised Medicine, TNO and University of Leiden, Utrechtseweg 48, 3700 AJ
Zeist and Leiden University, The Netherlands
The shift from a reductionistic towards a systems view is globally a key topic in Life Sciences. System-based
approaches need the understanding of the interconnectivity of systems and the organizing principles.
page 44 / 88
SB@NL2014 symposium
Symposium booklet
Moreover, it requires a shift from disease management to the domain of health promotion. In order to
achieve preventive strategies new insights need to be gathered on the concept of health and the measures
of health. Moreover, the diagnostic principles need to be further refined and needs to address the
personalized aspect. Diagnosis and in particular the dynamic relationship between symptoms play a key for
understanding systems and for modelling of health and disease. In Chinese medicine a holistic and
personalized approach is key for both diagnosis and intervention for diseases or support of health and this
creates a perfect match with Systems Science acting as a bridge between Western and Chinese medicine,
with the opportunity to understand the system regulation at the biochemical level. Moreover, based on a
system diagnostic perspective also the synergetic nature of multi component mixtures can be better
understood and will stimulate the network biology approaches as currently been developed in systems
pharmacology.
Haanstra, Jurgen
Targeting pathogen metabolism without collateral damage in the host
Jurgen R. Haanstra (1,2); Albert Gerding (1); Hermann-Georg Holzhütter (3); Balázs Szöör (4); Keith
Matthews (4); Jacky L. Snoep (2,5,6); Hans V. Westerhoff (2,6,7); Barbara M. Bakker (1,2)
(1) University of Groningen, University Medical Centre Groningen, Department of Pediatrics, Centre for Liver, Digestive
and Metabolic Diseases, Groningen, The Netherlands; (2) Department of Molecular Cell Physiology, Faculty of Earth
and Life Sciences, VU University, Amsterdam, The Netherlands; (3) Charité - Universitätsmedizin Berlin, Institut für
Biochemie, Berlin, Germany; (4) Department of Biochemistry, Stellenbosch University, South Africa; (6) School of
Chemical Engineering and Analytical Science, University of Manchester, Manchester, UK; (7) Swammerdam Institute
for Life Sciences, Faculty of Science, University of Amsterdam, Amsterdam, The Netherlands.
A huge challenge in the combat of infectious diseases is to target the disease-causing agent without
harming its host. One approach is to find unique drug targets in the pathogen-specific proteome, but in
eukaryotic pathogens those may be limited or the existing ones may not control a process that is vital for
pathogen survival. Therefore we should expand the search for drug targets by elucidating network-based
differences that convey a stronger effect of a drug in the pathogen than in the host.
Trypanosoma brucei is a eukaryotic parasite causing deadly diseases in human and cattle and that lives in
the serum of the mammalian bloodstream. There, T. brucei can only generate ATP in glycolysis, making
glycolysis a potent target pathway for antitrypanosomal drugs. Combining mathematical modelling and
wet-lab experiments we have previously identified glucose transport as the enzyme with the highest flux
control over trypanosome glycolysis. Indeed, glucose transport inhibitors killed trypanosomes. We found
that parasites that were able to survive this treatment, started to rewire carbon metabolism. Unexpectedly,
this adaptation even strengthened the potential of glucose transport as a drug target: glycolytic enzyme
expression was decreased and targetable antigens appeared on its membrane. These anti-homeostatic
adaptive responses of the parasite thus even lower the overall median lethal dose - or LD50 - of the drug.
But the challenge remains to combat the parasite inside the host. To address this we started with a
comparison with its closest neighbour in the host: the red blood cell. Comparison of two similarly detailed
kinetic models of glycolysis of the trypanosome and of the erythrocyte revealed that inhibitors of glucose
transport that are competitive for glucose affect trypanosome glycolysis much stronger than glycolysis in
the erythrocyte. We have developed co-culture experiments between trypanosomes and human
erythrocytes that enables testing of differential effects. The experiments reveal that, even in a situation
that is more favourable to the trypanosome, glucose transport inhibitors can selectively kill the
trypanosome without affecting survival or metabolism of erythrocytes.
We thus conclude that selective killing of trypanosomes in the context of the host is possible even when we
hit a target shared by the parasite and its host. This exemplifies that the arsenal of potential selective drug
page 45 / 88
SB@NL2014 symposium
Symposium booklet
targets can be broadened beyond the pathogen-specific proteome through a differential network-based
approach integrating in silico modelling and in vitro/in vivo experimentation.
Hoen, Peter-Bram ‘t
Invited Speaker
Novel insights in the regulation of mRNA transcription, processing and translation through
integration of mRNA sequencing data
de Klerk, Eleonora (1,*); Anvar, Seyed Yahya (1,2,*); Fokkema, Ivo F.A.C. (1); Vermaat, Martijn (1,2);
Thiadens, Klaske A.M.H. (3); von Lindern, Marieke (3); Turner, Stephen W. (4); den Dunnen, Johan T. (1,2); ‘t
Hoen,. Peter A.C. (1)
(1) Department of Human Genetics and (2) Leiden Genome Technology Centre, Leiden University Medical Centre,
Leiden, 2300 RC, The Netherlands; (3) Department of Haematopoiesis, Sanquin Research and Landsteiner Laboratory,
AMC/UvA, Amsterdam, The Netherlands; (4) Pacific Biosciences, 1380 Willow Road, Menlo Park, CA 94025, USA. (*)
equal contributions.
To date, the human transcriptome is known to contain around 80,000 protein-coding transcripts, and the
estimated number of proteins synthesized range from 250,000 to 1 million. All these transcripts and
proteins are coded by less than 20,000 genes, suggesting extensive regulation at transcriptional, posttranscriptional and translational level. I will discuss how integration of data obtained from diverse RNA
sequencing technologies (RNA-seq, deepCAGE, ribosome footprinting) improves our understanding of these
regulatory mechanisms and I will illustrate how these mechanisms jointly orchestrate the changes in
protein demands during muscle differentiation. The individual regulatory layers appear to be tightly linked,
with extensive cross-talk and feedback between them. To decipher the cross-talk between transcriptional
and posttranscriptional regulation, we analysed PacBio® single-molecule long sequencing reads capturing
full-length mRNA molecules. These data show that the vast number of potential combinations between
alternative transcription start sites, alternatively spliced exons and alternative polyadenylation sites result
in a relatively limited number of mRNA species, supporting the tight coupling between these processes.
Further integration of RNA sequencing data will elucidate the true complexity of the transcriptome and its
multi-layered regulation.
Hof, Wim van den
A systems biology approach in primary mouse hepatocytes unravels mechanisms of Cyclosporin
A-induced hepatotoxicity
Van den Hof, Wim (1,4); Van Summeren, Anke (1,4); Lommen, Arjen (2,4); Coonen, Maarten (1,4); Brauers,
Karen (1); van Herwijnen, Marcel (1); Wodzig, Will (3,4); Kleinjans, Jos (1,4)
(1) Department of Toxicogenomics, Maastricht University, Maastricht, the Netherlands; (2) RIKILT, Institute of Food
Safety, Wageningen University and Research Centre, Wageningen, the Netherlands; (3) Department of Clinical
Chemistry, Maastricht University Medical Centre, Maastricht, the Netherlands; (4) Netherlands Toxicogenomics Centre,
Maastricht, the Netherlands.
The liver is responsible for drug metabolism and drug-induced hepatotoxicity is the most frequent reason
for drug withdrawal, indicating that better pre-clinical toxicity tests are needed. In order to bypass animal
models for toxicity screening, we exposed primary mouse hepatocytes for exploring the prototypical
page 46 / 88
SB@NL2014 symposium
Symposium booklet
hepatotoxicant cyclosporin A. To elucidate the mechanisms underlying cyclosporin A-induced
hepatotoxicity in a systems biology approach, we analysed expression levels of proteins, mRNAs,
microRNAs and metabolites. Integrative analysis of transcriptomics and proteomics showed that protein
disulfide isomerase family A, member 4 was up-regulated on both the protein level and mRNA level. This
protein is involved in protein folding and secretion in the endoplasmic reticulum. Furthermore, the
microRNA mmu-miR-182-5p which is predicted to interact with the mRNA of this protein, was also
differentially expressed, further emphasizing endoplasmic reticulum stress as important event in druginduced toxicity. To further investigate the interaction between the significantly expressed proteins, a
network was created including genes and microRNAs known to interact with these proteins and this
network was used to visualize the experimental data. In total 6 clusters could be distinguished which
appeared to be involved in several toxicity related processes, including alteration of protein folding and
secretion in the endoplasmic reticulum. Metabonomic analyses resulted in 5 differentially expressed
metabolites, indicative of an altered glucose, lipid and cholesterol homeostasis which can be related to
cholestasis. Single analyses of transcriptomics, proteomics and metabonomics and a systems biology
approach both reveal mechanisms underlying cyclosporin A-induced cholestasis demonstrating that
endoplasmic reticulum stress and the unfolded protein response are important processes in drug-induced
liver toxicity.
Hooft, Rob
Data at the Dutch Techcentre for Life Sciences
Boiten, Jan-Willem (1); Bonino, Luiz Olavo (2); Bouwman, Jildau (3,4); Eijssen, Lars (5); Evelo, Chris (5);
Finkers, Richard (6,7); Francissen, Femke (2); van Gelder, Celia (2,8); Groenen, Martien (6); Hooft, Rob
(2,11); Kok, Ruben (2); Mons, Barend (2,9)
1. CTMM-TraIT, Eindhoven; 2. Dutch Techcentre for Life Sciences, Utrecht; 3. TNO, Zeist; 4. Netherlands Metabolomics
Centre, Leiden; 5. BiGCaT, Maastricht University; 6. WUR, Wageningen; 7. Plant Research International, Wageningen;
8 Radboudumc, Nijmegen; 9. LUMC, Leiden; 10. SURFsara, Amsterdam; 11. Netherlands eScience Centre, Amsterdam;
12. UMCG, Groningen.
The Dutch Techcentre for Life Sciences (DTL) brings together the experts in technology supporting life
scientists. Its programmes are Data, Technology and Learning. DTL is formed by experts from many
academic institutes as well as companies. DTL Data has an international assignment: The DTL Data
programme hosts the Dutch node in ELIXIR, the European Research infrastructure for life science data.
ELIXIR NL works closely together with the Dutch representations in other Life Science ESFRI infrastructures
to make sure data handling expertise is shared. It is essential that close ties are knit also with Dutch
scientists involved in the setup of ISBE, the systems biology research infrastructure. The Netherlands'
assignment in the international ELIXIR organisation is currently focused on three areas: Data
Interoperability, computing and network infrastructure, and education. Nationally ELIXIR-NL makes sure all
Dutch Life Scientists know about and have access to all internationally offered resources, and that other
Dutch resources deemed valuable for international audience can be brought to a broader audience. DTL
Data brings life scientists and data experts together: DTL Data can establish contact between life scientists
and the right people to help fill their data stewardship plans; with expertise anywhere from highperformance computing to experiment planning and modelling. DTL Data also organizes and hosts regular
experience sharing meetings of project leaders in projects per life science sectors. We have regular
programmer meetings for sharing experience with techniques and applications. We also assemble people
with similar interest in "focus meetings" that can lead to DTL facilitated interest groups and working groups
tackling technological problems together.
page 47 / 88
SB@NL2014 symposium
Hornung, Bastian
Symposium booklet
Genomic and functional analysis of Romboutsia Ilealis CRIBT reveals adaptation to the small
intestine
Hornung, Bastian V.H. (1,3, *); Gerritsen, Jacoline (1,2, *); Renckens, Bernadet (4); Van Hijum, Sacha A. F. T.
(4,5); Martins dos Santos, Vitor A.P. (3); Rijkers, Ger T. (6,7); Schaap, Peter J. (3); de Vos, Willem M. (1,8);
Smidt, Hauke (1).
(1) Laboratory of Microbiology, Wageningen University, Wageningen, The Netherlands; (2) Winclove Probiotics,
Amsterdam, The Netherlands; (3) Laboratory of Systems and Synthetic Biology, Wageningen University, Wageningen,
The Netherlands; (4) Nijmegen Centre for Molecular Life Sciences, CMBI, Radboud UMC, Nijmegen, the Netherlands; (5)
NIZO Food Research, Ede, the Netherlands; (6) Laboratory for Medical Microbiology and Immunology, St. Antonius
Hospital, Nieuwegein, The Netherlands; (7) Department of Science, University College Roosevelt, Middelburg, The
Netherlands; (8) Departments of Microbiology and Immunology and Veterinary Biosciences, University of Helsinki,
Helsinki, Finland; (*) these authors contributed equally to this work.
Introduction and objectives
The microbial communities in the small intestine are dependent on their capacity to rapidly take up and
ferment available carbohydrates. To survive in a complex and highly competitive ecosystem, intestinal
microbes have adapted or even specialized in foraging certain niche-specific substrates. Recently the novel
species Romboutsia ilealis CRIBT was isolated from the small intestinal tract of rats. To gain more insight in
the metabolic and functional capabilities of this natural and abundant inhabitant of the small intestine and
its adaptation to the small intestine, a genome-guided physiological analysis was performed.
Materials and methods
The complete genome of R. ilealis CRIBT was sequenced and annotated, followed by a whole-genome
transcriptome analysis after growth on specific carbohydrates (glucose, L-fucose and fructo-oligosaccharide
[scFos]).
Results and discussion
R. ilealis CRIBT possesses a single, circular chromosome of 2.581.778 bp that contains 2351 predicted
proteins, and a non-mobilisable plasmid of 6.145 bp carrying eight predicted proteins. Analysis of the
genome revealed only limited ability to synthesize amino acids and vitamins. However, multiple and
partially redundant pathways for the utilization of a wide array of simple carbohydrates were identified.
The whole-genome transcriptome analysis allowed pinpointing components of the key pathways involved
in the degradation of glucose, L-fucose and scFos. These analyses revealed that R. ilealis CRIBT is a flexible
anaerobe that is adapted to a nutrient-rich environment in which carbohydrates and exogenous amino
acids and vitamins are abundantly available. Other features of ecological interest include urease and bile
salt hydrolase encoding genes. This work shows how a combination of genome mining and functional
analyses with single microbes can provide an overall insight in the genetic and functional potential of
specific members of the intestinal microbiota.
Horst, Rob ter
Adaptive value of metabolic responses to gene knockouts in yeast
Ter Horst, Rob (1); Papp, Balázs (2); Notebaart, Richard (1)
page 48 / 88
SB@NL2014 symposium
Symposium booklet
1. Radboud Institute for Molecular Life Sciences, Centre for Bioinformatics and Systems Biology, Radboud University
Medical Centre, Nijmegen, the Netherlands; 2. Synthetic and Systems Biology Unit, Institute of Biochemistry, Biological
Research Centre of the Hungarian Academy of Sciences, Szeged, Hungary
Background
Cells respond to new genetic mutations in a multitude of ways. It is often assumed that these changes are
adaptive, helping the cell's long-term survival. However, recent evidence suggests that the initial response
to large changes is not necessarily adaptive, and might even be maladaptive (i.e., have adverse effects).
Here we focus on the metabolic responses in yeast, and investigate whether the initial responses to
metabolic knock-outs (KOs) show (mal)adaptive signatures.
Methods
Constraint-based modelling (CBM) is a powerful tool for gaining insights into cellular metabolism. Using a
yeast metabolic network and a combination of two well-known CBM-techniques, flux balance analysis (FBA)
and flux variability analysis (FVA), we predict which transcriptional responses to metabolic knockouts
should provide adaptive value. We compare these predictions to the experimentally measured
transcriptional changes that take place in the first few generations after gene KO.
Results
Most transcriptional changes after gene KO appear to be fairly random, and do not correlate to the
predicted "optimal" changes. We did however also observe several interesting feedback mechanisms. Some
of these increase expression of isoenzymes (enzymes catalysing the same reaction) of the KO gene. This
might help restore reaction speeds towards their old wild-type levels. However, other feedback
mechanisms increase expression of genes in the same linear reaction pathway as the knockout gene. Since
the pathways are interrupted, these are actually maladaptive changes, possibly leading to a toxic build-up
of metabolites.
Conclusion
The cell's initial metabolic response to KO generally has no strong adaptive signature. The feedback
mechanisms that do exist are sometimes adaptive, but are just as likely to be maladaptive. Other data
shows a similar pattern for changes outside of metabolism.
Jacobsen, Annika
Modelling Wnt/β-catenin signalling
Jacobsen, Annika (1,2); Verkaar, Folkert (2,3); Heringa, Jaap (1,2); Smit, Martine J. (2,3); Feenstra, K. Anton
(1,2)
(1) Centre for Integrative Bioinformatics (IBIVU), Department of Computer Science, VU University Amsterdam; (2)
Amsterdam Institute for Molecules, Medicines and Systems (AIMMS); (3) Division of Medicinal Chemistry, VU
University Amsterdam
The Wnt/β-catenin signalling pathway is important for cell renewal, proliferation and differentiation. Active
Wnt/β-catenin signalling caused by specific mutations can play a role in oncogenesis. A better
understanding of these signalling mechanism is therefore crucial for efficient prevention and treatment.
We have constructed a Petri net model of Wnt/β-catenin signalling to capture the behaviour under
different mutations. The model is based on current knowledge from literature and experimentally obtained
data and consists of the core pathway proteins and their interactions. Simulations of the model with
different perturbations reproduces experimentally known responses of nuclear β-catenin. Our model can
be used to explain which perturbations cause active Wnt signalling and the underlying mechanisms. We are
now extending the model with genes that are known to activate Wnt signalling. This will hopefully lead to
page 49 / 88
SB@NL2014 symposium
Symposium booklet
greater understanding of how oncogenic signals propagate through the network from disrupting the
Wnt/β-catenin signalling pathway to transcription of target genes and regulation of gene expression.
Kelder, Thomas
Tackling systems complexity through data integration
Thomas Kelder (1); Georg Summer (2,3); Marijana Radonjic (1)
(1) EdgeLeap B.V., Utrecht, The Netherlands; (2) Maastricht University, CARIM, Maastricht, The Netherlands; (3) TNO,
Microbiology & Systems Biology, Zeist, The Netherlands
Life today is becoming increasingly data intense. Growing technological advances in molecular profiling
assays and self-monitoring devices stimulate innovative approaches to health management. The availability
of information on our lifestyle and health status (e.g. patient health records, longitudinal clinical and
physiological parameters, genetics and genomics information, energy expenditure records, dietary habits,
questionnaires outcomes, food purchase records etc.) provides an opportunity to integrate multiple
aspects of health and behaviour into a unified, systems, and person-specific health profile. On a personal
level, this allows making better informed decisions on how to stay healthy. On a population level, this
enables discovery of patterns which improve decisions on lifestyle and drug intervention strategies. The
challenge that remains is translation of this complex and diverse data into relevant knowledge to empower
such decision making. We develop and apply network-based platforms for integration and mining of diverse
data and knowledge to facilitate development of systems strategies for health improvement and
maintenance. For instance, this means organizing readouts of molecular assays such as tissue
transcriptome, clinical parameters, gut microbiome composition or personal 23andMe data with the
growing body of knowledge on molecular and physiological interactions. This enables “vertical” integration
of different levels of systems complexity to fully characterize the system under investigation, allows placing
personalised data in the context of population-based models and makes optimal use of own and existing
data and knowledge. Together, the ability to organize, integrate and exploit diverse and abundant
information on health-relevant interactions supports development of improved strategies to stay healthy
and combat disease.
Klein Entink, Rinke H.
Personalized inferences: the possibilities of N-of-1 and do-it-yourself trials over randomized
control studies.
Klein Entink, Rinke H. (1); Boessen, Ruud (1); Rubingh, Carina (1); Bijlsma, Sabina (1); Boorsma, André (1);
Clabbers, Nard (1)
TNO
The current trend towards personalized medical treatments and lifestyle advice is also evident in the field
of food and nutrition. The evidence base for treatment advice is usually based on (double blind)
randomized controlled trials (RCT), a classical, reliable method inherited from the pharmaceutical industry.
There are limitations to the use of RCTs for studying health effects of food products, however. Firstly,
because it is hard to hide what a person is eating. Secondly, an RCT provides evidence for the studied
page 50 / 88
SB@NL2014 symposium
Symposium booklet
population in general, while there may be substantial person-to-person variability in the effect size of a
treatment. For example, the stimulating effect of caffeine may vary substantially. Therefore, personalized
advice requires also a personalized study set-up. We will show the possibilities of N-of-1 trials and do-ityourself study designs to go towards personalized inferences and advice for healthy food and nutrition.
Kuiper, Esther
A standardised database of curated metabolic models for facilitated reconstruction and
comparative analysis of genome scale metabolic models
Kuiper, Esther (1); Molenaar, Douwe (1)
Systems Bioinformatics, VU University Amsterdam.
The sequencing of genomes has become easier and cheaper. The accumulation of these genomes still asks
for a closer examination in order to be of use in the understanding of the organism’s physiology. Metabolic
network reconstruction and the simulation of these networks are important tools for studying the systems
biology of metabolism and are, therefore, important next steps after the sequencing of the genome of
interest. While sequencing is getting easier, metabolic model reconstruction still takes a lot of time and
expert knowledge. The automation of this process will help in future studies to easily gain a good working
metabolic model of the species of interest. In this study we are creating a toolset for the improvement of
automatic reconstruction of metabolic models. Our aim is to create a database of curated models that is
easily accessible and reliable in providing where data came from. The database will connect models from
different sources, giving it a good coverage of metabolic reactions across a wide range of organisms. The
notation of information within these models is far from uniform and names for identical reactions across
organisms are often inconsistent. By assembling data from different sources these inconsistencies will
accumulate. Therefore we try to unite the reaction and species names as much as possible by mapping to
unique identifiers and notations, making use of the vast amounts of knowledge of enzymes, pathways and
chemical compounds that is currently available in online databases. Combining this information, from
curated metabolic models and available databases, generates an unambiguous database of sequencecomplex-reaction links. Making it possible to easily identify identical reactions in models from different
sources independent of the name spaces used in those models.
Kuzniar, Arnold
PIQMIe proteomics web server applied to dissect the mechanism of cancer therapy based on
interference with DNA repair
Kuzniar, Arnold; Zelensky, Alex; Kanaar, Roland
Erasmus MC Cancer Institute, Departments of Genetics and Radiation Oncology, Cancer Genomics Netherlands,
Rotterdam, The Netherlands
Stable isotope labelling with amino acids in cell culture (SILAC)-based mass spectrometry is a mature semiquantitative proteomics technique, whereby tens of thousands of peptides and thousands of (nonredundant) proteins are reliably identified and quantified in a typical unbiased proteome experiment. We
present the Proteomics Identifications and Quantitation’s Data Management and Integration Service or
page 51 / 88
SB@NL2014 symposium
Symposium booklet
PIQMIe that aids in reliable and scalable data management, analysis and visualization of semi-quantitative
proteomics experiments. PIQMIe readily integrates peptide and protein identifications and quantitations
from multiple experiments with additional biological information on the protein entries from UniProtKB,
and makes the linked data available in the form of a light-weight database, which enables dedicated
analyses and user-driven queries (Fig. 1A). Using the web interface, users are presented with a concise
summary of their proteomics experiments in numerical and graphical forms, as well as with a searchable
protein grid and interactive visualisation tools to aid in the rapid assessment of the experiments and in the
identification of proteins of interest. The web server also supports programmatic access through RESTful
web service. We used PIQMIe to mine and visualize the data from triple SILAC inverse labelling experiments
which revealed new DNA repair relevant integrators of the BRCA2 protein (Fig. 1B). These integrators were
identified by applying hyperthermia, an anti-cancer treatment used in the clinic to augment radio- or
chemotherapy. The results not only provide insight into DNA repair in response to radiation and DNA
damage based chemotherapy, but also identify changes in the BRCA2 protein complex that allow the
dissection of the molecular mechanism of how hyperthermia augments cancer treatment.
Reference
Kuzniar, A. and Kanaar, R. (2014) PIQMIe: a web server for semi-quantitative proteomics data management
and analysis, Nucleic Acids Res 42, W100-W106.
Landeweerd, Laurens
Invited Speaker
Visions from science: translating societal values of systems biology
Landeweerd, Laurens; Swierstra, Tsjalling; Van Driel, Roel
page 52 / 88
SB@NL2014 symposium
Symposium booklet
(1) Radboud University Nijmegen, FNWI, ISIS; (2) TU Delft, department of Biotechnology, Section BTS.
Landeweerd will provide for a sketch of the interrelation between science and society over the past 40
years, from curiosity driven research, to interaction with industry (1970s) and societal engagement and
societal criticism (1980s-1990s), to an integration of science, society and industry (2000s). He will discuss
the changing nature of research autonomy in the light of science-industry-interfaces and the need for
societal responsiveness. Three roles are central for normative reflection on new and emerging science and
technology: embedding research and innovation in society, critical observation of research and technology
and addressing current and future needs. The first of these is the impression most experts hold of
‘bioethics’, but the field also holds potential for increasing the values of science and for the scienceindustry-interface through the second and third role. To be able to better align science with societal
reflection, one needs to depart from a vocabulary of constraints and threats to a vocabulary of vision and
chances s. This also means that one needs to go beyond mere risk assessment. By the example of
Monsanto on the one hand, and the human genome project on the other, Landeweerd will discuss which
specific societal drivers and hurdles can be expected in relation to systems biology in the light of the
tension between fundamental and applied research.
Martines, Anne-Claire
How the interplay between long-chain fatty-acid influx, short-chain oxidation and mitochondrial
respiration leads to overload of the mitochondrial fatty-acid beta-oxidation
Martines, Anne-Claire M.F. (1); Van Eunen, Karen (1); Bakker, Barbara M. (1)
(1) Department of Paediatrics, Centre for Liver, Digestive and Metabolic Diseases, University of Groningen, University
Medical Centre Groningen, Antonius Deusinglaan 1, 9713 AV Groningen, The Netherlands
Recently, we showed by dynamic modelling that the mitochondrial fatty-acid beta-oxidation pathway (see
figure) is susceptible to substrate overload. At a high influx of palmitoyl-CoA, the steady-state flux dropped
and metabolites accumulated (van Eunen et al., 2013). Here, we investigate why the flux declines and how
the accumulation of intermediate CoA-esters depends on the kinetic parameters around the flux optimum.
We applied metabolic control analysis to the dynamic model to quantify how fluxes and metabolite
concentrations depend on the kinetic parameters. At the substrate overload point the steady-state flux
declined steeply and the carnitine O-palmitoyltransferase (CPT)1-catalyzed reaction gained negative flux
control. Positive flux control shifted to the C4- and C6 reactions by the medium- and short-chain acyl-CoA
dehydrogenases (MCAD and SCAD) and the medium-chain ketoacyl-CoA thiolase (MCKAT). We proved that
in the transition, the thermodynamic hurdle at medium/short-chain hydroxyacyl-CoA dehydrogenase
(M/SCHAD; Q/Keq = 1; Keq = 0.0002) - in combination with the low affinity of MCAD and SCAD towards
their substrates compared to their products - led to full occupancy of MCAD and SCAD by their short-chain
products. Consequently, CoA-esters accumulated and the steady-state flux dropped. In the hypothetical
case of a higher affinity of MCAD and SCAD for their short-chain substrates, the occupancy by their shortchain substrates was higher. This prevented the accumulation of CoA-esters and protected against the
overload phenotype. An alternative rescue was to shift the M/SCHAD equilibrium by increasing the
NAD+/NADH ratio, indicative of enhanced mitochondrial respiration. We conclude that the affinity of
MCAD and SCAD for their short-chain substrates is important for the prevention of substrate overload and
that the tight interplay between beta-oxidation and mitochondrial respiration runs via the short-chain
reactions catalysed by M/SCHAD, MCAD and SCAD.
page 53 / 88
SB@NL2014 symposium
Symposium booklet
Mondeel, D.G.A.
Making Systems Biology Work: Roadmap to Multiscale Predictive Modelling
Mondeel, D.G.A.; Westerhoff, H.V.; Barberis, Matteo.
SILS, University of Amsterdam
A current challenge in systems biology is the need for multiscale models covering a broad range of
biological functions and scales. This requires the integration of the hierarchical layers of regulation and
control present in the cell. This project aims to increase our understanding of bodily function in terms of
dynamically interacting modules in space and time. A complete systems understanding of the cell requires
integrated, multiscale modelling of the hierarchical layers of regulation and control. We aim to design such
a multiscale model focused on the cell cycle. We especially intend to incorporate interacting modules of
metabolism to be able to link metabolic impairments to dynamic consequences for the cell cycle.
Muyzer, Gerard
Invited Speaker
Microbial systems ecology - a systems biology approach to unravel the mysteries of microbial
communities
Muyzer, Gerard
Microbial Systems Ecology, Department of Aquatic Microbiology, Institute for Biodiversity and Ecosystem Dynamics
(IBED), University of Amsterdam, Amsterdam, The Netherlands
page 54 / 88
SB@NL2014 symposium
Symposium booklet
Bacteria are everywhere on earth present in large numbers. However, little is known of these bacteria,
because they are small and difficult to identify. It is well known that less than 1% of all bacteria in nature
have been isolated in pure culture (Alain and Querellou, 2009). So, basic questions, such as (i) who is there?
(ii) what are they doing? and (iii) how is a microbial community changing after perturbations, are difficult to
answer with traditional microbiological techniques. To answer these questions we have to use molecular
biological techniques. By using these techniques, such as cloning of PCR-amplified 16S rRNA genes, we now
know that microbial diversity is enormous and that most of the microorganisms are novel. However, these
techniques are limited in the number of samples that can be analysed and therefore statistical analysis is
difficult. Like in many biological disciplines Next Generation Sequencing (NGS) has also revolutionised the
field of microbial ecology. NGS can be used to detect all genes of all organisms in an environmental sample,
the so-called ‘metagenome’, at once. Moreover, NGS can be used to study gene expression of
microorganisms in microbial communities, an approach known as metatranscriptomics. This enables us to
study microbial communities using a systems biology approach (Raes and Bork, 2008).
In this lecture I will present the first results on the use of comparative genomics and metagenomics to
study microbial communities from soda lakes. Soda lakes are extreme environments with pH values
between 9 and 11, and salinities up to saturation. However, despite these extreme conditions, soda lakes
are highly productive and harbour diverse microbial communities (Sorokin et al., 2014). The sulfur cycle,
driven by sulfur-oxidising and sulfidogenic bacteria, is one of the most active element cycles in these
habitats. Members of the genus Thioalkalivibrio have versatile metabolic capabilities, including sulfide
oxidation, denitrification and thiocyanate utilisation. We have isolated more than 70 strains, for which the
genomes have been sequenced by the Joint Genome Institute of the U.S. Department of Energy. In
addition, the microbial communities from soda lake brines and sediments have been sequenced. The
availability of this sequence data will allow us to gain insight into the diversity of these bacteria, their niche
differentiation, and the molecular mechanisms by which they adapt to the extreme halo-alkaline
conditions. For this we will use a systems biology approach, combining different 'omics' techniques with
physiological experiments under well-defined conditions, and mathematical modelling. The results of these
experiments are of paramount importance, both for a basic understanding of life under extreme
conditions, as well as for the use of these bacteria in the sustainable removal of noxious sulfur compounds
from polluted environments.
References
Alian K. and J. Querellou (2009) Cultivating the uncultured: limits, advances and future challenges.
Extremophiles 13: 583-594.
Raes J. and P. Bork (2008) Molecular eco-systems biology: towards an understanding of community
function. Nature Rev. Microbiol. 6: 693-699.
Sorokin D.Y., T. Berben, E.D. Melton, L. Overmars, C.D. Vavourakis, and G. Muyzer (2014) Microbial diversity
and biogeochemical cycling in soda lakes. Extremophiles 18: 791-809.
Nevedomskaya, Ekaterina
Androgen Receptor profiling in tumour specimens yields hallmarks of prostate cancer outcome
Nevedomskaya, Ekaterina (1,6); Stelloo, Suzan (1); van der Poel, Henk G (2); de Jong, Jeroen (3); van
Leenders, Geert JLH (4); Jenster, Guido (5); Wessels, Lodewyk FA (6); Bergman, Andre M (7); Zwart, Wilbert
(1).
Division of Molecular Pathology (1), Urology (2), Pathology (3), Molecular Carcinogenesis (6) and Medical Oncology (7),
the Netherlands Cancer Institute, Amsterdam, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands; Department of
Pathology (4) and Urology (5), Josephine Nefkens Institute, Erasmus Medical Centre, Rotterdam, The Netherlands.
page 55 / 88
SB@NL2014 symposium
Symposium booklet
Prostate cancer (PC) is the most common malignancy in men and one of the leading causes of cancerrelated deaths in the Western world. While in many cases the PC can be treated with curative intent, there
is no treatment for recurring disease, which occurs in 30% of the patients. Biomarkers for outcome are
urgently needed in order to identify high-risk patients, who can benefit from additional therapy and have a
chance for cure. Androgen Receptor (AR) is a critical driver at all stages of PC progression, even after the
development of resistance to AR-targeting androgen deprivation therapy. To identify prognostic markers
and to determine causal players in prostate cancer progression, we turned to the level on which Androgen
Receptor acts: protein-DNA binding and transcriptional regulation. We performed genome-wide profiling of
chromatin accessibility and AR/DNA interaction at different stages of prostate cancer progression. We
identified a distinct Androgen Receptor/chromatin binding profile between primary prostate cancers and
tumours with an acquired resistance to therapy. We further performed multiple levels of integration of this
genomic data with other datasets: gene expression in perturbed cell lines, clinical gene expression and
survival data. This allowed us dissecting the functionality and clinical relevance of genes affected by the
observed change of transcriptional regulation. As a result we propose a concise gene signature with strong
prognostic potential. The power of the proposed approach lies in the integration of multiple datastreams
on transcriptional regulation and functional gene expression, which has a potential to provide biologically
relevant prognostic signature. This innovative pipeline for biomarker discovery can be easily implemented
in other fields of oncology.
Nordholt, Niclas
Characterisation of an IPTG-inducible promoter in Bacillus subtilis
Nordholt, Niclas (1); van Heerden, Johan (1); Bruggeman, Frank J. (1)
(1) Systems Bioinformatics, VU University Amsterdam.
Here we characterise Bacillus subtilis strains harbouring a genomic insertion of a fluorescent protein under
the control of an IPTG-inducible promoter. We measure the fluorescence and optical density of an
exponentially-growing cell culture, using a microplate reader. We assess the dependency of promoter
activity as function of the IPTG concentration across a range of growth conditions. Using basic
page 56 / 88
SB@NL2014 symposium
Symposium booklet
microbiological growth theory, we explain why promoter activity is dependent on growth conditions.
Finally, we present general guidelines for the use of microplate readers to study microbial growth. We
shortly discuss pitfalls and limitations of this widely-used technique. These guidelines should help to
standardise growth-rate determinations across different labs.
Oertlin, Christian
MiPaSt: a PathVisio plugin for the analysis of regulatory interactions
Christian Oertlin, Martina Kutmon, Susan LM Coort, Lars MT Eijssen
Department of Bioinformatics - BiGCaT, Maastricht University
PathVisio is an open source tool for pathway visualisation and analysis (www.pathvisio.org). Besides editing
pathways, it allows the user to load experimental expression data to visualise them on the pathway
diagram and to perform pathway statistics. Pathway statistics is used to obtain overrepresented pathways,
for example using the WikiPathways pathway collection (www.wikipathways.org). Recently, PathVisio has
been extended with functionality to enhance the study of regulatory mechanisms in biological processes,
such as microRNA regulation. The Regulatory Interaction plugin was created to be able to upload an
interaction file and visualise the expression of interaction partners of a selected gene or protein in a side
panel.
When using PathVisio to perform statistics on expression data, regulatory interaction information cannot
be included in the ranking of pathways. For that reason we are currently developing a new plugin, MiPaSt
(miRNA Pathway Statistics) that combines the core statistics function of PathVisio with the Regulatory
Interaction plugin to extend the pathway statistics options, focusing on miRNA-gene interactions.
The MiPaSt plugin uses miRNA and gene expression data together to identify relevant pathways. Known
miRNA-target interactions, loaded through the Regulatory Interaction plugin, allow users to define the
regulatory interplay between miRNAs and genes. As an example, the user could specify that a
downregulated gene should only be considered when a known microRNA regulator is upregulated, or the
other way around.
MiPaSt is a PathVisio plugin that provides a pipeline to integrate, analyse and visualise microRNA and gene
expression together in biological pathways. The plugin ranks the pathways based on expression levels of
the genes but also of their regulators. The MiPaSt plugin enables the user to predict the regulation of
pathways by the effects of miRNA on their target genes.
Ommen, Gert-Jan
Invited Speaker
BBMRI and beyond: the role of biobanks in the integration of high-dimensional data in human
biology and health care
Ommen, Gert-Jan B. (1,2).
(1) Department of Human Genetics, Centre for Medical Systems Biology, Leiden University Medical Centre (LUMC); (2)
Biobanking and Biomolecular Research Infrastructure (BBMRI-NL).
Unravelling the complex interplay of genes and environment to achieve better treatment and prevention in
health care requires very large datasets. To achieve statistical robustness, even small subsets need to have
page 57 / 88
SB@NL2014 symposium
Symposium booklet
sufficient size for proper retrospective and prognostic value. In the first decade of this century, initiatives
were deployed to connect research of sample-and datasets across international borders and to establish
distributed biobanking research infrastructures. In Europe, the first transnational project was
GenomEUtwin. The first global initiative was P3G, the Public Project in Population Genomics, a
Canadian/European initiative, and in 2008 the Austrian-led European biobanking infrastructure BBMRI
(Biobanking and BioMolecular Resources Research Infrastructure) started its preparatory phase funded by
the ESFRI action. From 2008-2011 this grew into a pan-European consortium with 52 participants and >200
associated members in 30 member states, which in turn led to the establishment of a large number of
National Nodes, starting with BBMRI-NL and BBMRI.se. Between 2011 and 2013 BBMRI transitioned into
the BBMRI-ERIC (BBMRI European Research Infrastructure Consortium), formally established in 2013 with
its seat in Graz. BBMRI-ERIC presently counts 12 member states and 5 observers and is in size and scope of
participating organisations the largest European research infrastructure. The first transnational sample-and
data sharing project BBMRI-LPC (BBMRI Large population cohorts) started in 2013 and, through an open
transnational call for proposals in mid-2014 BBMRI-LPC, is now on its way to provide international research
groups centrally-funded access to samples and data enrichment from 20 large European population
cohorts. In the future, to consolidate high-dimensional data utilisation in the life sciences, further
integration of European Research infrastructures is foreseen, to be driven by ELIXIR, and with BBMRI-ERIC
in a co-leading role.
Overmars, Lex
Exploring the pan-genome of the haloalkaliphilic sulfur-oxidizing bacterium Thioalkalivibrio
Overmars, L (1); Muyzer, G (1)
(1) Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam, Amsterdam, Netherlands
Thioalkalivibrio are chemolithoautothrophic sulfur-oxidizing bacteria growing at a pH of 10 and salt
concentrations up to 4.3 M of sodium. An insight into their genomic potential and diversity can help to
obtain a better understanding of the molecular mechanisms by which Thioalkalivibrio adapts to these
extreme conditions. We have sequenced the genomes of 70 strains of Thioalkalivibrio. These strains were
isolated from soda lakes in Mongolia, Siberia, California, Egypt and Kenya, which vary in salinity and brine
chemistry. Comparative analysis revealed that the pan-genome consists of more than 10,000 orthologous
groups (OGs), of which ~15% make up the conserved core. The core genome is mainly composed of
housekeeping, (core-) metabolism- and information storage/processing- related genes, whereas the
accessory genome is characterized by an overabundance of genes involved in signal transduction and cell
wall biogenesis. We have clustered the strains based on presence/absence of OGs and found two major
clusters consisting of isolates from the Asiatic (Siberia and Mongolia) and the African (Kenya and Egypt)
continents, respectively. We are currently extending our analysis by focusing on pathways and proteins
potentially important for the adaptation to their environment to learn more about how Thioalkalivibrio can
excel in these harsh conditions of high pH and salinity.
Paalvast, Yared
A systems biology approach points to increased FFA influx as main contributor of LXR-agonist
induced hepatic triglyceride accumulation
page 58 / 88
SB@NL2014 symposium
Symposium booklet
Paalvast, Yared (1); Hijmans, Brenda (1); Tiemann, Christian (3,4); Grefhorst, Aldo(6); Boesjes, Marije(1);
Van Dijk, Theo (2); Tietge, Uwe (1); Kuipers, Folkert (1,2); Van Riel, Natal (3,4); Groen, Albert (1,2,4,5);
Oosterveer, Maailke (1)
Departments of (1) Pediatrics and (2) Laboratory Medicine, University of Groningen, University Medical Centre
Groningen, The Netherlands; (3) Department of Biomedical Engineering, Eindhoven University of Technology,
Eindhoven, The Netherlands; (4) Netherlands Consortium for Systems Biology, University of Amsterdam, Amsterdam,
The Netherlands; (5) Groningen Centre for Systems Biology, University of Groningen, Groningen, The Netherlands; (6)
Department of Internal Medicine, Erasmus MC, Rotterdam, The Netherlands.
Liver X receptor (LXR) agonists have shown promise in the treatment of atherosclerosis. Unfortunately, a
severe side-effect of LXR-agonist treatment is accumulation of triglyceride (TG) in the liver, hampering its
usability in a clinical setting. To better understand LXR-agonist induced hepatic TG accumulation we
employed a computational modelling approach called Analysis of Dynamic Adaptations in Parameter
Trajectories (ADAPT). ADAPT integrates experimentally obtained concentration and flux measurements
with a model of ordinary differential equations (ODE). By using a longitudinal dataset, ADAPT predicts how
parameters in the ODE model must change through time in order to comply with this dataset. The
advantage of ADAPT over reasoning alone is that predictions necessarily take into account all factors in a
model simultaneously and can therefore also expose behaviour that is counterintuitive. In this study, we
show that LXR-activation induces an increase in both input and output fluxes to hepatic TG and that hepatic
steatosis results from only a minor imbalance between the two. It is generally believed that LXR-induced
hepatic steatosis results from increased de novo lipogenesis (DNL). In contrast, ADAPT predicts that
especially in the early phase of LXR-activation, influx of free fatty acids (FFA) is the major contributor to
hepatic TG accumulation. For validation, we measured the contribution of plasma FFA to hepatic TG by
infusing 13-C labelled palmitate in C57Bl/6J mice treated for one day with the LXR-agonist T0901317
followed by mass-isotopomer distribution analysis. In agreement with ADAPTs prediction, there was a 5fold increase in the contribution of plasma palmitate to hepatic oleate and palmitoleate. In conclusion, this
study shows that hepatic TG accumulation in the early phase of LXR-agonist treatment is caused by
increased FFA influx rather than increased DNL.
Peyriéras, Nadine
Invited Speaker
Modelling multilevel dynamics in animal embryonic morphogenesis from multiscale imaging
data
Peyriéras, Nadine.
BioEmergences USR, Centre National de la Recherche Scientifique (CNRS), Gif sur Yvette, France.
We approach the understanding of animal model organisms’ embryonic morphogenesis through the
quantitative analysis of multiscale in vivo imaging data. The cellular level of organisation is taken as
resulting from the integration of sub-cellular and supra-cellular processes. Cell dynamics are investigated
through 3D+time imaging of developing embryos with fluorescent nuclear and membrane staining. The
automated reconstruction of the cell lineage tree, annotated with nucleus and membrane segmentation,
provides measurements for cell behaviour: displacement, division, shape and contact changes, as well as
fate and identity. This quantitative data is sufficient to find statistical models for cell proliferation and cell
descriptors evolution in time and space, and characterise the spatial and temporal length scale of cell
displacements and tissue deformations. Confronting numerical simulation derived from a multi-agent
based biomechanical model with empirical measurements extracted from the reconstructed digital
specimens, is the basis for testing hypotheses for processes underlying early embryogenesis. Further
page 59 / 88
SB@NL2014 symposium
Symposium booklet
correlating cell behaviour, tissue biomechanics and biochemical activities by comparing the patterns
revealed by cell fate, velocity, strains or gene expression, is a step toward the integration of multi-level
dynamics. This overall framework lays the ground for a transdisciplinary approach of living systems’
morphogenesis.
Piebes, Diewertje
Transcription repression dynamics
Diewertje G.E. Piebes(1); Hermannus Kempe(1); Frank J. Bruggeman(2); Gijsbert J. van Belle(3); Adriaan B.
Houtsmuller (3); Pernette J. Verschure*(1)
(1) Swammerdam Institute for Life Sciences, University of Amsterdam, Science Park 904, 1098 XH Amsterdam, the
Netherlands; (2) Systems Bioinformatics, Amsterdam Institute for Molecules, Medicines and Systems, VU University
Amsterdam, De Boelelaan 1085, 1081 HV, Amsterdam, The Netherlands; (3) Department of Pathology, ErasmusMC,
Rotterdam, The Netherlands.
Gene expression is affected by the complex interplay of proteins that directly or indirectly bind to DNA and
alter the chromatin composition. Transcription factors and chromatin binding proteins allow gene
expression on/off switching. Since the binding and recruitment of transcriptional activators and silencers is
connected it is difficult to study these actions separately in an endogenous context. In the present study,
we aim to understand the dynamics of transcriptional repression and to identify the role of a transcriptional
repressed chromatin composition on the transcription repression kinetics. We used a mammalian cell line
equipped with a genomically integrated reporter gene array containing lac and tet operator DNA binding
sequences, a CFP reporter gene and bacteriophage MS2 RNA hairpin loops. This reporter gene array allows
us to locally change the chromatin structure, to interfere with the transcriptional activity of the reporter
gene and to measure MS2 transcription repression dynamics in real-time. We determined the MS2
transcription dynamics quantifying YFP tagged MS2 binding protein levels at the reporter gene array upon
releasing tet operator targeted transcriptional activator (VP16). Moreover, we compared MS2 transcription
repression dynamics when the reporter gene array is in a transcriptionally-inactive context upon targeting
Methyl-CpG-binding protein-2 (MeCP2) thereby inducing recruitment of transcriptional repressor proteins
including histone deacetylases. Our data show that the decline of reporter gene activity is time-consuming
compared to the rapid release of the transcriptional activator. The reporter gene activity declined faster
when the chromatin is in a transcriptional repressed composition due to MeCP2 targeting. We conclude
that releasing a transcriptional activator decreases the rate of transcription although gene activity lingers
long after its disappearance.
Rabbers, Iraes
Does Escherichia coli express its H+-ATPase to an optimal level?
Rabbers, Iraes (1); Schwabe, Anne (1); Bruggeman, Frank (1)
(1) VU University Amsterdam, department of Systems Bioinformatics.
When nutrients are in excess, natural selection maximizes the specific growth rate of a microorganism. The
specific growth rate equals the rate of the synthesis of a new cell per unit cell, and is set by the identity and
level of the expressed enzymes. Since resources are limited, the expression of a specific protein generally
page 60 / 88
SB@NL2014 symposium
Symposium booklet
occurs at the expense of other proteins, which amounts to a protein cost (shown by Dong & Kurland, 1995
for instance). At the same time, protein expression confers a benefit if the protein contributes to fitness.
The state of maximal fitness is reached when all proteins contribute to fitness and are at precisely tuned
levels, such that none of the proteins is expressed at a too high or too low level (Berkhout et al., 2013). Our
aim is to understand how capable wild type Escherichia coli is in achieving optimality across a range of
environments. To achieve this we study whether E. coli is able to optimise the level of H+-ATPase across
conditions. Through a combination of modelling and experiments, we hope to gain a better understanding
of the evolutionary processes taking place in nutrient adaptations.
Reijnders, Maarten
Bioinformatics and systems biology integration for algal biotechnology
Maarten J.M.F. Reijnders (1); Ruben G.A. van Heck (1); Carolyn M.C. Lam (1); Vitor A.P. Martins dos Santos
(1,2); Peter J. Schaap (1)
(1) Laboratory of Systems and Synthetic Biology, Wageningen University, Dreijenplein 10, Building number 316; (2)
LifeGlimmer GmbH, Markelstrasse 38, 12163 Berlin, Germany.
Many species of microalgae can be used as industrial biosynthesis platforms for biofuels and high value
products. However, due to genetic constraints they are not yet competitive against non-renewable
alternatives. Metabolic engineering approaches will help in improving their productivity, but many genes
and metabolic pathways remain unknown. Recent advances in bioinformatics and systems-biology,
together with an increased amount of algal omics data, are providing the means to address this.
Improvements in algal annotations together with systems-biology modelling approaches of the metabolic
and regulatory networks are essential first steps towards engineering industrially valuable algal strains [1].
We have developed a consensus-based protein function prediction method for annotation of polyphyletic
species such as microalgae. Annotations retrieved by this method can be used for (genome-scale)
modelling of microalgae to refine our understanding of the capabilities of a specific alga and to provide a
basis for applications in biotechnology.
Reference
[1] Reijnders, M.J.M.F., and et al., Green genes: bioinformatics and systems-biology innovations drive algal
biotechnology. Trends in Biotechnology, 2014. 32(12): p. 617-626.
Figure 1: A multidisciplinary workflow integrating bioinformatics, systems biology, and metabolic
engineering/synthetic biology of microalgae. Black arrow, in silico data or predictions; white arrow,
experimental (wet-lab) data.
page 61 / 88
SB@NL2014 symposium
Rens, Lisanne
Symposium booklet
Mechanical cell-ECM interactions amplifies response to substrate stretch and induces cell
alignment
Rens, Lisanne (1); Merks, Roeland (2)
(1) Life Sciences group, Centrum Wiskunde en Informatica, Amsterdam, The Netherlands; (2) Mathematical Institute,
Leiden University, Leiden, The Netherlands
It is crucial for tissue engineered constructs to be able to function properly within the body. For this reason,
cells in the engineered tissue should be organized according to the tissue that is to be replaced. Guiding cell
organization is not trivial, as many factors contribute to cell migration. The mechanical properties of the
extracellular matrix (ECM) influences their behaviour. It has been observed that cells on a stretched
substrate elongate and orient in the direction of stretch. Mechanical cues in the ECM not only serve as a
guidance for cells, but also as a way for cells to communicate with neighbouring cells by exerting forces on
the ECM. Thus, self-organisation of cells into patterns likely depends on the interactions between cells and
the ECM. To study these interactions, we have established a computational model that describes cells
generating and responding to strains in the substrate. We simulate this model using a hybrid Cellular Potts
model of cell motility and a Finite Element Model of the ECM [1]. We assume that cells preferentially
adhere to higher strains in the substrate. This model reproduces the observation that cells orient in parallel
to stretch. Further, we found interesting implications of cell traction forces. The model suggests that when
cells generate strains in the matrix, cell orientation is quicker, more accurate and aligned strings of cells
form in parallel to stretch, as shown in figure 1. This organisation has been observed [2]. We conclude that
by pulling on the substrate, cells can not only better coordinate their own behaviour in accordance to the
external mechanical signal, but also promote this behaviour in neighbouring cells. Thus, cellular traction
page 62 / 88
SB@NL2014 symposium
Symposium booklet
forces can amplify external mechanical signals in the extracellular matrix to elevate cellular selforganisation in line with the external mechanical cue.
Rienksma, Rienk
Selected Speaker
Systems biology of intracellular Mycobacteria
Rienksma, Rienk A. (1); Suarez Diez, Maria (1); Schaap, Peter J. (1); dos Santos, Vitor A.P. (1).
(1) Laboratory of Systems and Synthetic Biology, Wageningen University and Research Centre, Dreijenplein 10, 6703
HB Wageningen, the Netherlands.
Mycobacterium tuberculosis (Mtb) is the causative agent of tuberculosis (TB), an infectious disease that is a
serious health threat in developing countries. An estimated one-third of the world population is infected
with TB and multidrug resistant, extensively drug resistant, and totally drug resistant strains have emerged.
Therefore, novel therapeutics and intervention strategies are required. Approximately one-fourth of the
Mtb genome contains genes that encode proteins directly involved in its metabolism. These enzymes
provide a potential source of drug targets to effectively disrupt metabolic functioning. Systems-level
metabolic network reconstructions and the derived constraint-based (CB) metabolic models are efficient
tools to probe mycobacterial metabolism. We have created the most comprehensive model of Mtb
metabolism to date and show that it outperforms its predecessors in both flux and gene essentiality
predictions. CB metabolic models rely not only on a high quality reconstructed metabolic network, but also
on a suitable objective function. An important assumption of CB metabolic models is that optimization
principles underpin metabolic states (the total of fluxes through metabolism). It is assumed that a cell "tries
to achieve a metabolic objective". However, the objective of Mtb in the host is unclear. So-called condition
specific models are aimed at maximizing the agreement between relatively easily obtainable data, such as
gene expression data, and flux, thereby circumventing the necessity of suitable objective functions that
describe the behaviour of Mtb in the host. Our comprehensive transcriptome of intracellular mycobacteria
was used to formulate condition specific objective functions for both Mtb and host, enabling the systematic
probing of combinatorial drug targets and elucidation of potential "escape routes" after in-silico application
of drugs with known metabolic targets.
Rozendaal, Yvonne
Dynamical modelling of the onset and progression of the metabolic syndrome through
longitudinal data integration using ADAPT
Rozendaal, Yvonne (1); Wang, Yanan (2,3); Willems van Dijk, Ko (2,4); Rensen, Patrick (2); Groen, Bert (3);
Hilbers, Peter (1); van Riel, Natal (1)
(1) Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven; (2) Department of
Endocrinology and Einthoven Laboratory for Experimental Vascular Medicine, Leiden University Medical Centre,
Leiden; (3) Department of Pediatrics/Laboratory Medicine, University Medical Centre Groningen, Groningen; (4)
Department of Human Genetics, Leiden University Medical Centre, Leiden.
The metabolic syndrome (MetS) is a cluster of comorbidities including obesity, hyperglycemia, dyslipidemia
and hypertension. Due to its heterogeneous metabolic origin and clinical presentation, it is a challenging
task to unravel which pathways are responsible for the onset and progression of MetS. By exploiting a
page 63 / 88
SB@NL2014 symposium
Symposium booklet
dynamic modelling approach, we aim to identify underlying mechanisms in an individualised manner. The
onset of MetS is studied in male APOE*3Leiden.CETP mice through diet-induced obesity (high-fat,
cholesterol-rich diets), resulting in insulin resistance and dyslipidemia. Data of body weight, plasma
metabolites and liver lipids are integrated in a mathematical model of glucose and lipid metabolism and the
dynamic adaptations in the network are analysed using ADAPT (Analysis of Dynamic Adaptations in
Parameter Trajectories) [1]. The longitudinal change in body weight and plasma metabolites differs
significantly among different diets, but also the heterogeneity among mice fed the same diet is large (see
figure). Conventional data analysis based on statistical measures excludes extreme phenotypes as these are
classified as outliers. Since these extreme phenotypes may be of particular interest, we performed the data
analysis using an individualised approach through data integration in the model using ADAPT. The change in
metabolic health, as monitored by glucose tolerance and dyslipidemia, of the individual mouse depends
largely on the diet, while taking into account the negative relation between insulin sensitivity and body
weight. These results reveal that high-fat diets induce obesity, hyperglycemia, impaired glucose tolerance
and insulin resistance, whereas additional dietary cholesterol induces hyperlipidemia. The individualised
data analysis approach supports these findings, and moreover, is able to take heterogeneity (due to
differences in food intake) in metabolic health into account.
[1] Tiemann et al. PLoS Comput Biol. 2013 9(8).
Sips, Fianne
page 64 / 88
SB@NL2014 symposium
Symposium booklet
Modelling gastric bypass–induced improvement of glycaemic control following a meal
Sips, Fianne (1); Jansen, Mattijs (1); Snel, Roderick (1); Hilbers, Peter (1); Van Riel, Natal (1)
(1) Eindhoven University of Technology, Department of Biomedical Engineering
Gastric bypass surgery has been shown consistently to result in improved glycaemic control in obese, Type
II Diabetic subjects. The mechanisms causing this marked effect include profound changes of gastrointestinal physiology, gut hormone secretion, and bodyweight. As the surgery affects the gastro-intestinal
system directly, the physiology of -and response to- the ingestion of a meal are particularly perturbed.
Although it is clear that postprandial glucose and insulin levels are better controlled post-surgery than in
pre-surgery subjects, this effect is not well understood as it is a result of a dynamic interplay of multiple
mechanisms. Quantifying the contributions of each of these components to the improvement in glycemic
control thus remains a challenge. In order to untangle this response, to characterise the improved glucose
tolerance of the post-surgery subject and to provide an integrated view of gastric bypass improvement
mechanisms we propose a mathematical model-based method. We applied a dynamic model to describe
the meal response of morbidly obese subjects undergoing Roux-en-Y gastric bypass. Measurements of the
meal response of obese controls, non-diabetic subjects undergoing bypass surgery and diabetic subjects
undergoing bypass surgery (both before and after surgery) were obtained. The data included glucose meal
rate of appearance, in addition to gut hormones kinetics. We report preliminary results of the model
analysis, in which we examine whether the model is able to incorporate the observed changes of GLP-1
kinetics and glucose rate of appearance, as both are known to change markedly following gastric bypass.
The model is then used to analyse the transition of glucose control from the pre- to post-gastric bypass
surgery situation.
Smith, Robert
Decoupling activity from activation: shifting phytochrome signals away from red light
Smith, Robert (1); Samodelov, Sophia (2); Pel, Eran (1,3); Borst, Jan Willem (3); Zurbriggen, Matias (2);
Fleck, Christian (1)
(1) Laboratory of Systems & Synthetic Biology, Wageningen UR, Netherlands; (2) Synthetic Signalling Networks,
University of Freiburg, Germany; (3) Laboratory of Biochemistry, Wageningen UR, Netherlands.
Synthetic biologists aim to engineer tools that can be used uniformly across biological systems to achieve
desired responses. Out of this research, the field of optogenetics has emerged using light-regulated
synthetic networks to control biological systems. Recent constructs have used plant photoreceptors to
control cellular mechanisms, such as transcription, protein localization and hormone concentrations. The
system we use in this study consists of the red/far-red photoreceptor phytochrome B (phyB) and its
interaction partner PHYTOCHROME INTERACTING FACTOR 6 (PIF6). Under red light, a chromophore
attached to phyB forms the active Pfr state allowing phyB to interact with PIF6 and regulate transcription.
Under far-red light illumination, the chromophore reverts back to the inactive Pr state preventing phyBPIF6 interactions and downstream processes. Thus, both in synthetic systems – that use the phyB (1-650)
protein fragment – and in planta, phyB activity occurs specifically when exposed to red light. In this study,
we decouple absorption of light by phyB and wavelength-dependent activity of phyB. Experimentally, we
measured absorption spectra of phyB (1-650) and a transcriptional readout (SEAP) across the light
spectrum. Using this information, we constructed a mathematical model of the system. From this,
sensitivity analysis determines which model parameters can be manipulated to produce shifts in phyB
page 65 / 88
SB@NL2014 symposium
Symposium booklet
activity without changing light absorption properties. Using these methods, we find that a Nuclear
Exclusion Signal (NES) tag and limited availability of PIF6 generates a green shift in phyB (1-650) activity.
These model predictions are proven experimentally, showing that the biological network downstream of
light absorption plays a key role in coordinating phyB activity. The novel finding that phyB activity can be
green-shifted independent of light absorption suggests that evolution has constrained the phyB activity to
red light in plants.
Soons, Zita
Selected Speaker
The role of urea cycle activity in metabolic transformation of breast cancer
Soons, Zita (1,2); Mitra, Devina (3); Bernhardt, Stephan (3); Wolf ,Thomas (2); Korf, Ulrike (3); Wiemann,
Stefan (3); König, Rainer (2,4)
(1) Department of Knowledge Engineering, Maastricht University, Maastricht, The Netherlands; (2) Theoretical
Bioinformatics, German Cancer Research Centre, Heidelberg, Germany; (3) Molecular Genome Analysis, German
Cancer Research Centre, Heidelberg, Germany; (4) Leibniz Institute for Natural Product Research and Infection Biology,
Hans Knöll Institute, Jena, Germany.
Data from breast cancer tumours are heterogeneous with respect to the patients' characteristics and
treatment. Identification of different molecular subtypes in breast cancer based gene expression profiles
has led to improved patient stratification and refinement of therapeutic strategies. However, treatment of
the triple negative subtype still remains a challenge, amongst others due to heterogeneity within this
subtype. Although the mutations in individual cancer patients are different, we found that the metabolic
phenotypes can be comparable and are a good predictor of survival. Using the METABRIC dataset [1]
containing gene expression profiles of 2000 patients we identified nitrogen metabolism as a key metabolic
pathway deregulated in breast cancer. Based on these genes, we classified patients into metabolic subtypes
associated with good and bad outcome. Many cancers show an elevated uptake of glutamine, which
exceeds the cellular demands of biomass formation. By metabolic modelling using constraints on the
metabolic profiles from Jain et al. [2] and additional measurements from our lab, we found that glutamine
uptake in addition functions as a bioenergetics substrate through activity of the urea and the TCA cycle –
the cancer bicycle- to sustain its growth. We validated these findings in cell lines. In future we will carry out
perturbation experiments to investigate if metabolic vulnerabilities in the activity of the cancer bicycle may
be exploited to improve therapy of a subtype of triple negative breast cancer patients.
References
[1] Curtis et al., “The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel
subgroups”, Nature 486:346-352, 2012
[2] Jain et al., “Metabolite profiling identifies a key role for glycine in rapid cancer cell proliferation”,
Science 336: 1040-1044, 2012.
Souza, Terezinha
Carcinogen-induced transcriptomic alterations advance HepG2 towards hepatocellular
carcinoma
Souza, Terezinha (1); Jennen, Danye (1)l; Van Delft, Joost (1); Van Herwijnen, Marcel (1); Kyrtoupolos,
Soterios (2); Kleinjans, Jos (1).
page 66 / 88
SB@NL2014 symposium
Symposium booklet
(1) Department of Toxicogenomics, Maastricht University, Maastricht, 6229 ER, The Netherlands; (2) National Hellenic
Research Foundation, Institute of Biology, Medicinal Chemistry and Biotechnology, Athens,11635, Greece.
Hepatocellular carcinoma (HCC) is the main type of primary liver malignancy and due to its low survival
rate, accounts for a major fraction of global cancer-related deaths. Benzo(a)pyrene (BaP), an established
human carcinogen representative for the class of polycyclic aromatic hydrocarbons, is a ubiquitous toxin
present in high levels in cigarette smoke and in incomplete combustion products. Recently, this compound
was included as a risk factor for HCC development, but the exact underlying mechanisms remain unknown.
We have investigated the effects of BaP and its major metabolites, 3-OH-BP, 9,10-diol and BPDE on gene
expression modulation in hepatic cell models and evaluated their implication with regard to the onset of
HCC. Based on whole genome gene expression data, HepG2 was selected among three hepatic cell models
(HepaRG, HepG2 and PHH) as the most suitable for studying BaP-induced promotion towards a more HCCtype state. Exposure to BaP or to its metabolites for 6, 12 and 18 h resulted in major expression
modulations induced by BaP, BPDE and to a minor extent, by 9,10-diol, demonstrating activation of several
transcription factor networks, involved in carcinogenesis (AP-1, ATF-2 and HIF-1α), BaP metabolism (AhR)
or oxidative stress (Nrf2). BaP- and BPDE-exposed HepG2 cells shared the most molecular features with
HCC, with pathways related to established cancer hallmarks. Furthermore, data suggests that BaP, 9,10-diol
and BPDE may also influence HepG2 progression towards a more malignant phenotype, by deregulating the
expression of genes associated with poor prognosis in HCC patients.
Stigter, Hans
Identifiability of parameters in SB models - a novel algorithm
Stigter, Hans (1); Molenaar, Jaap (2)
(1)+(2): Biometris, Wageningen University, The Netherlands
Before a set of model parameters can be estimated from measured data, it is essential to know whether
the parameter estimation exercise is not doomed to fail because of a lack of (local) structural identifiability.
With this we mean that parameters often cannot be estimated from a data record, no matter how large the
data set is. Lack of identifiability is a problem of model structure combined with a poor measurement
strategy or experimental design.
In the literature many methods are available for testing a model for a lack of local structural identifiability.
All these methods, however, are computationally cumbersome and require possibly days of computation
time. The conclusion of these long computations is usually nothing more than a yes/no answer to the
question: ``is this model, together with the chosen sensor set, locally structural identifiable?'' Uncertainty
propagation and an estimate on the errors in the parameter estimates that have been obtained are often
not provided.
We present a novel algorithm that is based on a detailed analysis of parametric output sensitivities.
Application of a singular value decomposition to a parametric output sensitivity matrix quickly learns where
exactly strong parameter correlations are situated in the model equations. In a complementary analysis
these correlations are further investigated on the basis of a symbolic computation. The identifiability
analysis of large models can now be carried out in a substantial smaller amount of time. Moreover,
calculation of parametric output sensitivities also brings us closer to an analysis of the errors in the
parameter estimates (Fisher Information Matrix).
We will demonstrate our method through application of the novel identifiability test to several SB models
(e.g. Goldbeter, NFκ-B model). A lack of local structural identifiability is demonstrated for certain choices of
sensor combinations, thereby confirming literature results that were obtained on basis of long (symbolic)
computations.
page 67 / 88
SB@NL2014 symposium
Symposium booklet
Stolle, Sarah
Elucidating the effect of diet and exercise on the metabolic decline in ageing mouse livers by
metabolic modelling
Stolle, Sarah (1,3); Ciapaite, J. (1,3); Reijne, A.C. (1,3,4); Talarovicova, A. (1,3); Li, Y. (6); Van Dijk, G. (3,4,5);
Groen, A.K. (1,3); Reijngoud, D.J. (1,2,3); Rossell, S. (7); Bakker, B.M. (1,3)
(1) Department of Paediatrics, (2) Department of Laboratory Medicine, Centre for Liver, Digestive and Metabolic
Disease, University Medical Centre Groningen, University of Groningen, Groningen, The Netherlands; (3) Systems
Biology Centre for Energy Metabolism and Ageing , (4) Unit Neuroendocrinology, Centre for Behaviour and
Neurosciences, (5) Centre for Isotope Research, University of Groningen, Groningen, The Netherlands; (6) Department
of Genetics, University Medical Centre Groningen, University of Groningen, Groningen, The Netherlands; (7)
Computational Cancer Biology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The
Netherlands
Ageing is defined as a functional decline of an organism, thereby increasing the risk of death. The rate of
ageing depends on the genotype of an organism and on environmental influences to which the organism is
exposed. In this study we aim to understand the influence of diet and exercise on the decline in metabolic
function in mice. To study the metabolic decline we combine transcriptomics data with genome-scale
metabolic models. This enables us to create connections between the different hierarchies of the network.
The study design included six different conditions of mouse ageing: mice were given a high or low fat diet
and were additionally divided into ad libitum (control), running wheel or calorie restricted groups. We
follow the mice throughout their lifetime. At four time points (6, 12, 18 and 24 months) five mice per
cohort underwent in vivo flux measurements, where after they were sacrificed for further experiments.
Currently, we are measuring the transcriptome of liver, adipose tissue, heart and skeletal muscle by RNASeq for all conditions and time points. The genome-scale metabolic model is based on the iSS1393 mouse
model by Heinken et al (2013). First, we updated and extended the fatty-acid metabolism. Now, we are
building condition-specific models by integrating transcriptomics data, obtained from the animal
experiments, using the EXAMO method (Rossell et al, 2013). A comparison of the condition-specific models
will yield reactions and metabolites that become deregulated with age, and estimates at which time point
metabolic deregulation occurs. Our systems biology approach will give us insights into the relation between
age-associated metabolic decline and lifespan in mice. Ultimately, the goal is to translate this knowledge
into interventions that would benefit human healthy ageing.
Stuger, Rogier
Structural systems biology: the forces that compact and expand bacterial nucleoids
Rogier Stuger (1); Hans V. Westerhoff (1,2,3)
(1) Molecular Cell Physiology, VU University Amsterdam; (2) Synthetic Systems Biology, University of Amsterdam; (3)
MCISB, University of Manchester
Bacteria condense their chromosomes into nucleoids without tools like histones and nucleosomes.
Proposed nucleoid-compaction forces include DNA supercoiling, macromolecular crowding, and interaction
with various DNA-binding proteins. Transertion (coupled transcription, translation, protein translocation)
page 68 / 88
SB@NL2014 symposium
Symposium booklet
links the chromosomes to the plasma membrane, which counteracts nucleoid compaction. We
experimentally tuned DNA supercoiling and transertion in intact cells, and conclude that nucleoid
compactness is mainly determined by the balance of crowding and transertion, with a minor contribution
from DNA supercoiling. To explain our results, we extended a mathematical model that links supercoiling,
crowding, and other compaction forces to nucleoid volume by adding redistribution of twist and writhe by
DNA-binding proteins and nucleoid expansion by transertion.
Suarez-Diez, Maria
Deciphering Mycobacterium tuberculosis infections through the integration of heterogeneous
molecular networks and datasets
Suarez-Diez, Maria (1); Rienksma, Rienk A. (1); van Dam, Jesse CJ (1); Schaap, Peter J. (1); Martins dos
Santos, Vitor A.P. (1)
Laboratory of Systems and Synthetic Biology, Wageningen University, The Netherlands
Current biology research generates an ever-increasing deluge of omics derived data, each pertaining to a
single level of the biological system. A major challenge in Systems Biology is to develop methods to
combine the information at the metabolic and overarching regulatory levels. Methods have been
developed to infer regulatory networks and to integrate them into consensus sets. However, each set is
partly lost when building a consensus network. We have developed a methodology to generate coexpression networks and we propose an integration framework where the networks are kept and analysed
to combine the information from each one. We have analysed the co-expression networks of Mycobacteria
tuberculosis, the causative agent of tuberculosis, generated using over 600 expression experiments and
have studied the regulation of some key process allowing the bacteria to adapt to low oxygen and metal
availabilities (1). To explore the bacterial metabolism we have generated a genome-scale constraint-based
metabolic model of M tuberculosis, partly based on previous models. We have quantitatively validated it
with 13C measurements (2). To understand the onset of infection, we have used dual RNA-seq to
simultaneously analyse the transcriptomes of host and pathogen. During infection, M. tuberculosis resists
page 69 / 88
SB@NL2014 symposium
Symposium booklet
the harsh environment of phagosomes. We have combined the knowledge extracted from the coexpression networks and the metabolic model with expression data. Our results confirm the importance of
the mycobacterial pathways for cholesterol degradation and iron acquisition during infection and our
analysis shows that their regulations are closely interlinked.
References
(1) van Dam et al (2014). Integration of Heterogeneous Molecular Networks to Unravel Gene-regulation in
Mycobacterium Tuberculosis. BMC Systems Biology
(2) Rienksma et al Systems-level modelling of mycobacterial metabolism for the identification of new
(multi-)drug targets. Seminars in Immunology in press.
Summer, Georg
cyNeo4j - connecting Neo4j and Cytoscape
Summer, Georg (1,2); Kelder, Thomas (3); Ono, Keiichiro (4); Heymans, Stephane (1); Demchak, Barry (4).
(1) CARIM, Maastricht University, Maastricht, The Netherlands; (2) TNO, Zeist, The Netherlands; (3) EdgeLeap BV,
Utrecht, The Netherlands; (4) Department of Medicine, University of California San Diego, La Jolla, USA.
Neo4j provides a platform to store large scale graph data and run complex computations to analyse the
stored graph. It only provides basic visualization capabilities in comparison to Cytoscape, which excels in
these tasks. Integration of Cytoscape and Neo4J functionalities will provide a perfect platform to improve
both computational analysis and manual exploration and interpretation. To this end, we developed cyNeo4j
- a Cytoscape app that allows the user to connect to a Neo4j server and transfer a graph from Cytoscape to
Neo4j and back. This allows the execution of computationally demanding tasks on a Neo4j server within
Cytoscape. CyNeo4j is accompanied by Neo4j plugin implementations for the NetworkAnalyzer app and the
ForceAtlas2 layout to demonstrate this use case. The cyNeo4j app combined with the Neo4j plugins provide
an example of enhancing Cytoscape with a data and service provider like Neo4j.
Summer-Kutmon, Martina
Multi-omics data analysis using pathway and network approaches
Summer-Kutmon, Martina (1,2); Coort, Susan L (1); Evelo, Chris T (1,2)
(1) Department of Bioinformatics - BiGCaT, NUTRIM School for Nutrition, Toxicology and Metabolism, Maastricht
University, The Netherlands; (2) Maastricht Centre for Systems Biology (MaCSBio), Maastricht University, The
Netherlands.
The advances in high-throughput measurement and computational technologies enable researchers to look
beyond single biological processes and start investigating interactions on a system wide level. Pathway
diagrams and biological networks are powerful tools to integrate different kinds of experimental data
(transcriptomics, proteomics, metabolomics, fluxomics) with existing knowledge from pathway and
interaction databases. The open-source and widely-used pathway and network visualization and analysis
tools PathVisio and Cytoscape provide, together and independently, many different approaches to study
complex biological data. The pathway editor and analysis and visualization tool PathVisio
(www.pathvisio.org) can be easily extended with new functionality through so-called plugins. Key features
of PathVisio are drawing pathway diagrams, integrating and visualizing multi-omics data on pathways,
page 70 / 88
SB@NL2014 symposium
Symposium booklet
providing additional information about the elements in the pathway from online databases, and performing
pathway statistics to identify relevant pathways. Cytoscape (www.cytoscape.org) is a widely-adopted
network analysis and visualization software. Cytoscape can be easily extended through apps and a
collection of more than 70 apps provides additional features for e.g. data visualization, data integration,
functional analysis, integrated analysis or network generation. Pathways are network-like in nature and the
WikiPathways app for Cytoscape also enables users to open PathVisio pathways in Cytoscape. They can
then be easily extended with e.g. regulatory information using the CyTargetLinker app. Here we would like
to demonstrate how PathVisio and Cytoscape can be used together to integrate, analyse and visualise
different types of data together.
Swierstra, Tsjalling
Invited Speaker
Responsible research and innovation: a new challenge to systems biology
Swierstra, Tsjalling (1).
(1) Department of Philosophy, Faculty of Arts and Social Sciences, Maastricht University (UM).
This lecture will introduce the concept op Responsible Research and Innovation (RRI) by explaining its four
key dimensions: anticipation, inclusion, reflectivity, and responsiveness. I will discuss some societal,
technoscientific and economic backgrounds that help explain RRI’s recent rise to prominence, and will try
to build a case that RRI is not a passing fad but rather reflects a structural change in the relationship
between science, technology, and society. The new social contract between science, technology, and
society, is marked by a greater emphasis on ethical, legal, and societal aspects and by an increased
page 71 / 88
SB@NL2014 symposium
Symposium booklet
emphasis on dialogue with stakeholders. Finally, I will argue that to gain and foster society’s trust,
biotechnologists need to develop new attitudes and skills.
Szklarczyk, Radek
Systems biology methods for metabolic disorders
Szklarczyk, Radek (1); Smeets, Bert (1)
(1) Maastricht University Medical Centre+
In genetic metabolic disorders approx. 30-50% of genome-wide sequencing studies find no causative
mutation that can be linked to disease phenotype. It is thus important to develop tools that allow
prioritization of patients' genetic variants in the context of metabolic disorders. I will present a number of
advanced computational approaches that focus on the system-level analyses of genes implicated in disease
pathways: 1) Sensitive homology detection methods to identify fast-evolving components of metabolic
systems, 2) Gene history-aware analyses of metabolism from their evolutionary origin to the current-day
human pathways, 3) Coexpression analyses of metabolic systems showing simultaneous activation or
simultaneous down-regulation of whole pathways. These methods allow selection of additional candidate
genes (for example, transporters, regulators, and novel enzymes) that are functionally associated with the
affected metabolic pathway. I will present examples of identified disease proteins, the computational clues
that lead us to discovery and experimental validations and characterizations of candidate genes. The results
directly lend itself to genetic diagnostic tests in patients with metabolic disorders, and open possibilities for
designing therapeutic interventions.
References
Tucker E, Wanschers B, Szklarczyk R, et al., Plos Genetics, PLoS Genet. 2014, 9: e1004034
Szklarczyk, R, Wanschers B, et al., Human Molecular Genetics 2013, 22: 656–667
Balsa E, et al., Cell Metabolism 2012, 16: 378–386
Szklarczyk R, Wanschers B et al., Genome Biology 2012, 13:R12
Theunissen, Daniel
RNA-seq data analysis using high performance computing.
Theunissen, Daniël (1), Coonen, Maarten (2) ; Suppers, Pascal (2); Jennen, Danyel (1); Kleinjans, Jos (1)
(1) Department of Toxicogenomics, Maastricht University, P.O. Box 616, 6200 MD, Maastricht, The Netherlands; (2)
Faculty Office FHML , Maastricht University, P.O. Box 616, 6200 MD, Maastricht, The Netherlands.
The use of High Performance Computing (HPC) is becoming more apparent in modern research. In
disciplines as quantum mechanics and high energy physics, HPC has been used for years. With recent
developments in Life Sciences Research – generation of large number of sequencing samples , protein
structure simulations, advanced MRI imaging etc. – this domain will also greatly benefit from HPC
infrastructures for compute-intensive tasks. A couple of years ago, SURFsara (commissioned by NBIC)
constructed the Life Sciences Grid (LSG), specifically dedicated for research groups working in the Life
Sciences domain. The LSG consists of several compute clusters that are placed at academic institutions
across the Netherlands. Since August 2014, the LSG-nodes for Maastricht University are available for use.
Currently, the department of toxicogenomics is using the local cluster that is set-up in Maastricht for the
page 72 / 88
SB@NL2014 symposium
Symposium booklet
analysis of high-throughput sequencing data. Because of the size and complexity of these sequencing data
a researcher cannot perform the data analysis of these samples on their own workstation. Therefore, a
solution is needed to increase the computational power to analyse these data. The local cluster in
Maastricht contains 128 CPU-cores and 512 GB of RAM, which is a major improvement over the locally
available hardware. The adjustments to the RNA-seq analysis pipeline that was already deployed were
minor and the pipeline was up and running within a single workday. This move allowed the researchers to
run four times as many samples as before. In the future the analysis will be ported for use at the complete
GRID infrastructure so 10,000 potential job slots become available for use. This will give the possibility for
running the analysis of hundreds of samples concurrently. Also for computationally very intensive tasks, like
integrating data for a systems biology approach, the GRID infrastructure can become an indispensable part
of the research infrastructure.
Thieme, Sebastian
rxncon compiler – Bridging large scale reconstruction of signalling networks and mathematical
modelling
Thieme, Sebastian (1); Rother, Magdalena (1); Krantz, Marcus (1)
1) Humboldt-Universität zu Berlin
Network reconstruction is a process to formalise an in silico description of a biological network. This is an
important step to analyse this network and gain new insights. There are well established reconstruction
methods for metabolic networks, but the reconstruction process for cellular signal transduction networks is
more difficult and less well established. In a signalling network, the components can undergo multiple
reactions. The combination of these possibilities results in the so called combinatorial complexity problem
and therefore in highly complicated networks. As networks get large, the mathematical equations fundamental for modelling and analysing - get more complicated and difficult to manage. One way to
handle this is to systematise and condense the system like it is done in rule-based or reaction contingency
based formats. For this purpose, we developed the rxncon framework. It is based on a reaction-contingency
description and supports large scale reconstruction of signalling networks. The rxncon language mirrors the
empirical data, but creates a gap to established modelling methods. To overcome this gap, we developed
the rxnconcompiler - a python library that interprets the rxncon language. The network reconstruction is
translated into and stored in an object orientated structure, which we export into the BioNetGen Language.
Additionally, the structure enables further processing of the data and export of the stored information into
different mathematical models. Hence, the rxnconcompiler provides a bridge between a scalable network
reconstruction formalism and mathematical modelling.
Thijssen, Bram
Selected Speaker
Understanding the variability in drug response in a panel of breast cancer cell lines using
computational models
Thijssen, Bram; Jastrzebski, Kathy; Beijersbergen, Roderick; Wessels, Lodewyk
Division of Molecular Carcinogenesis, Netherlands Cancer Institute
page 73 / 88
SB@NL2014 symposium
Symposium booklet
Cancer cell lines differ widely in their sensitivity to anticancer drugs. While many oncogenic drivers and
drug resistance mechanisms have been discovered, it is generally unclear how these mechanisms interact
in each cell line to make the cell line sensitive or resistant to a particular drug. We set out to explain this
variability in drug sensitivity in a panel of 30 breast cancer cell lines. We characterized these cell lines at the
DNA, RNA and protein level, and accurately measured the proliferation under treatment with various
different kinase inhibitors. We then constructed computational models encompassing several of the
important driver pathways and sensitivity mechanisms, and tested how well these models describe the
available data. After selecting the well-fitting models, we could use these models to estimate the relative
contribution of each of the interacting mechanisms to the proliferation of the cells under drug treatment.
For example, the models indicated that FGF2 autocrine signalling contributes to fast proliferation; that
SGK1 expression provides a bypass for Akt signalling in some cell lines; or that the expression level of 4EBP1 is a key determinant of mTOR-inhibitor sensitivity on top of other genetic alterations in the PI3Kpathway. Additionally, we could use these models to thoroughly analyse which parts of the data cannot be
explained. This greatly narrows down which follow-up studies are necessary to advance our understanding
of drug response. These results show that knowledge-based computational models can be used to
systematically study drug response in cell line panels. Such systematic, quantitative understanding of drug
response will be useful in working towards precision medicine for individual cancer patients.
Verschure, Pernette
Selected Speaker
Epigenetics makes us tick: Functional performance, time-delay, memory, cell-cell variability and
robustness
Kempe, Hermannus (1); Anink-Groenen, Lisette C. M. (1); Bruggeman, Frank J. (2); Piebes, Diewertje G. E.
(1); Vuist, lona M. (1); Verschure, Pernette J. (1)
(1) Swammerdam Institute for Life Sciences, University of Amsterdam, Science park 904, 1098 XH Amsterdam, The
Netherlands; (2) Systems Bioinformatics, Amsterdam Institute for Molecules, Medicines and Systems, VU University
Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands.
Epigenetic gene regulation is crucial for cell type specific gene expression in higher eukaryotes. Chemical
modifications of DNA/chromatin and chromatin conformational changes are mechanisms of epigenetic
regulation. Epigenetic gene regulation confers cellular phenotypic stability, while allowing changes in
response to environmental cues. Derangement in epigenetic gene regulation has severe effects on cell
behaviour and contributes to a diseased state.
Epigenetic regulation can be dynamic, occurring as (transient, stochastic) oscillations in different cells in a
clonal population. This dynamic nature explains the lack of conclusive understanding of epigenetic
regulation since most observations assume cells are in phase and are based on cell-population averages.
The crossing of temporal, spatial and organization scales, all in association, are a great case for fundamental
systems biology.
We aim to understand the complex molecular principles (such as time-delay, memory, cell-cell
variability and robustness) of epigenetic regulation underlying functional genomic performance that
ultimately determines phenotypic behaviour. We use computational simulation in combination with
engineered mammalian cell systems. This approach enables us to modulate the (epi)genetic state at a
single gene locus at predefined regions of the genome and measure input-output relationships of
(epi)genetic regulation in multiple cells in a population. We developed techniques allowing
systematic, quantitative measurements of changes in the transcription rate as function of the epigenetic
chromatin state in single living mammalian cells with high precision, real-time and high-resolution
combined with stochastic modelling. Presently, in the context of a large EU H2020 consortium
(EpiPredict coordinated by PJV) we will focus on epigenetic state switching during the development
page 74 / 88
SB@NL2014 symposium
Symposium booklet
of breast cancer endocrine therapy resistance. Our approach of modelling and quantitative sampling
opens an unexplored field of research with great potential for medicine.
Wegrzyn, Agnieszka
The bigger picture – human peroxisomal single-enzyme deficiencies in fatty-acid metabolism
studied through genome-scale models.
Wegrzyn, Agnieszka (1); Suárez Diez, María (2); Martins dos Santos, Vitor A.P. (2); Bakker, Barbara (1)
(1) Systems Biology Centre for Energy Metabolism and Ageing and Department of Paediatrics, Centre for Liver,
Digestive and Metabolic Disease, University Medical Centre Groningen, University of Groningen, Groningen, The
Netherlands; (2) Systems and Synthetic Biology, Wageningen University, Agrotechnology & Food Sciences,
Wageningen, The Netherlands
The majority of the fatty acids (FAs) present in the human body are oxidized via the mitochondrial βoxidation pathway. Very-long chain FAs, phytanic acid, dicarboxylic acids and bile-acid precursors cannot,
however, be oxidized by mitochondria. Their metabolism is dependent on peroxisomal α- and β-oxidation,
and supported by ω-oxidation in the endoplasmic reticulum. Peroxisomal fatty-acid oxidation (FAO), in
contrast to the mitochondrial FAO, produces H2O2 instead of FADH2 and is only indirectly linked to the
mitochondrial respiratory chain via an NADH redox shuttle. While peroxisomal FAO may result in a loss of
ATP when compared to mitochondrial FAO, its crucial role as a detoxifying pathway is stressed by the
existence of several inborn peroxisomal deficiencies. Despite the severe phenotype of many peroxisomal
disorders, we don’t know enough about the regulation of FA metabolism and compartmentalisation to
develop strategies for treatment or diagnosis. The aim of this project is to increase our understanding of
peroxisomal FA metabolism and the associated diseases. We use systems biology tools, including genomescale and detailed dynamic models, to study the peroxisomal FA metabolism. The genome-scale metabolic
model is based on the newest version of Recon 2, the human metabolic reconstruction by Thiele et al.
(2013). We curated and extended the FAO pathways to reflect the current state of knowledge. Flux balance
analysis is used to determine the flux distribution in the core pathways of FA metabolism and Flux
Variability Analysis is used to determine the robustness of the network. With single-gene knockout analysis
we predicted how the flux distribution changes in human Refsum disease (FA α-oxidation defect). In this
project, a systems biology approach is applied to study the systems response to defects in FAO. We present
the curated and extended fatty-acid oxidation module and show initial predictions of the effect of Refsum
disease on the metabolic network.
Weijers, Dolf
Invited Speaker
Genetic control of growth and pattern formation in the early plant embryo
Weijers, Dolf (1)
(1) Laboratory of Biochemistry, Wageningen University and Research Centre.
Multicellular organisms are composed of different cell types with unique properties. Precursors for each
tissue are generally formed early during embryogenesis, and growth and pattern formation processes
ensure the generation of an ordered, functional arrangement of the appropriate number of cells. A key
question is how these are coordinated in space and time to generate shape and function. We address this
page 75 / 88
SB@NL2014 symposium
Symposium booklet
question in the early plant embryo where, in the absence of cell migration, shape is chiefly determined by
oriented cell division. Hormones, notably auxin and cytokinin, play a central role in both growth and
patterning, but the mechanisms through which they act are largely unknown. I will present our recent work
aimed at identifying the transcriptional signatures of early embryonic cell types, and the cellular basis for
the establishment of multicellular patterns in 3D, as well as its genetic control.
Westerhoff, Hans
Invited Speaker
Integrative systems biology: avenues towards precision science, medicine and biotechnology
Westerhoff, H.V. (1,2,3); and friends.
(1) Synthetic Systems Biology, SILS, the University of Amsterdam; (2) Amsterdam Institute for Molecules, Medicines and
Systems, VU University Amsterdam; (3) Manchester Centre for Integrative Systems Biology, University of Manchester.
At critical points, science, engineering and medicine falter when addressing ecology, politics, the economy,
or disease. This happens when understanding requires more than three dimensions, more than the linear
superposition of one-dimensional thoughts, or more than the search for the culprit. Because they excel in
this irreducible complexity, addressing this issue for living systems may help manage the others. For this
we need a new, third type of science, which Occam did not foresee. Integrative systems biology
amalgamates ‘forces’ from entirely different dimensions into realistic models of how together they drive
functions. These models should be of sufficient complexity (‘and not simpler’). Because this complexity is
much upwards from 300 genes, this long seemed impossible, and it still is for any single laboratory or
average consortium. Now, the unique accomplishments of Systems Biology include breakthroughs through
this stalemate, with genome wide integration of experimental data enabling the computation of biological
(mal)function. Systems biology has delivered. It has delivered discoveries impossible to make with the
traditional life sciences, and with implications well beyond biology. But, there should be more to come…. I
will present a framework of avenues for Systems Biology that will integrate the activities of many scientists,
and thereby enable the resolution of multiple, otherwise irresolvable, societal and scientific issues. Fasten
your seat belts.
Wolstencroft, Katy
Invited Speaker
Making results reusable: the next five years
Wolstencroft, Katy (1); Stanford, Natalie (2); Owen, Stuart (2); Krebs, Olga(3); Nguyen, Quyen (3);
Golebiewski, Martin (3); Kania, Renate (3); Snoep, Jacky(2); Mueller, Wolfgang (3); Goble, Carole (2).
(1) Leiden University; (2) University of Manchester, (3) Heidelberg Institute for Theoretical Studies (HITS).
Systems Biology involves modelling dynamic processes in biological systems by integrating data and
knowledge. The integration of heterogeneous datasets is typically required for model construction; and
understanding the relationships between them is essential for interpreting modelling results. Annotating
data and models using common formats and vocabularies can facilitate this process, and repositories for
sharing and linking systems biology results enable wider discovery and reuse. The SEEK platform
(http://www.seek4science.org/), for example, provides infrastructure and methodologies for sharing
systems biology assets, such as data, models, protocols and results. It builds on community standards and
already supports large European Systems Biology consortia. The SEEK is part of a larger ecosystem of
page 76 / 88
SB@NL2014 symposium
Symposium booklet
supportive computational resources, which also include databases, standards, informatics tools and
modelling frameworks. The ISBE project (Infrastructure for Systems Biology in Europe
http://project.isbe.eu/) aims to bring these resources together into a shared infrastructure that will serve
the needs of the European community and leverage European expertise. Here, we review current resources
for finding and reusing Systems Biology assets and discuss how new and continuing initiatives will help to
promote sharing and reuse in the future.
Wopereis, Suzan
Ranges of phenotypic flexibility in 100 healthy subjects
Wopereis, Suzan; Bakker, Gertruud; De Jong-Rubingh, Carina; Dijk-Stroeve, Annelies; Van Ommen, Ben;
Hendriks, Henk; Stafleu, Annette; Van Erk, Marjan.
TNO, Utrechtseweg 48, 3704 HE, Zeist
In this study, we aimed to assess the ranges of phenotypic flexibility as a measurement of health within the
apparently healthy population. Therefore, a total of 100 healthy subjects were enrolled (50 males, 50
females). We included males and females with a range in age (from 20 to 70 years) and in body fat
percentage (low, medium, high), to ensure variation in phenotypic flexibility. Phenotypic flexibility was
quantified by measuring in each volunteer the response of 160 markers to the PhenFlex challenge test (PC).
The PC is a drink containing high amounts of fat and glucose. We have shown previously that this challenge
test is able to quantify the adaptive capacities of most relevant metabolic processes for diet-related health.
The markers were selected to monitor the response of the following 4 processes relevant for phenotypic
flexibility: glucose metabolism, lipid metabolism, protein metabolism, and low grade chronic stress. At t=0
(fasting) and 6 time-points (t= 0.5, 1, 2, 4, 6 and 8 h) after PC, blood was sampled from each subject to
measure the total of 160 markers related to glucose metabolism, lipid metabolism, protein metabolism and
low grade chronic stress. Next, the range in phenotypic flexibility of the study population was analysed in
the “health-space”, a tool developed at TNO that shows individual phenotypic flexibility in a 4 dimensional
space defined by the 4 metabolic processes. The health space showed a different adaptation to PC in the
extremes of the recruited population: persons of young age with low to normal fat percentage had a
significant different response to PC compared to persons of old age with normal to high fat percentage
(both genders). Furthermore, the health space allowed the quantification of the individual metabolic health
state. Ultimately, visualizing phenotypic flexibility in the health space before and after (nutritional)
interventions will allow the evaluation of individual efficacy of the intervention.
Wu, Wen; Van Sluijs, Bob; Hof, Kevin
Selected Speaker
BananaGuard: biocontrol of Fusarium oxysporum using Pseudomonas putida (iGEM-2014-WUR)
Van Rosmalen, Rik; De Koster, Walter; Van Sluijs, Bob; Hof, Kevin; Van Baalen, Jeremy. Herpers, Michiel;
Van der Ploeg, Max; Robert Finestra, Teresa; Birk, Marlene; Correa Marrero, Miguel; Marolt, Tjasa; Wu,
Wen
Laboratory of Systems and Synthetic Biology, Laboratory of Microbiology, Wageningen University and Research
Centre.
page 77 / 88
SB@NL2014 symposium
Symposium booklet
Fusarium species are known to infect a wide range of crops and cause large losses in agriculture. Our
project aims to use an engineered strain of the native soil bacterium Pseudomonas putida to protect
banana plants against Fusarium oxysporum infection. Upon sensing of fusaric acid excreted by F.
oxysporum, fungal growth inhibitors will be produced by P. putida to prevent infection of the banana
plants. To minimise the environmental impact of the fungal inhibitors to the soil microbiome, the bacteria
will contain a kill switch that will cause it to terminate once the threat of F. oxysporum to the bananas has
been alleviated. Furthermore, a double dependent plasmid system will be used to prevent the spread of
artificial genetic material to surrounding soil bacteria. In summary, we hope to develop a showcase for the
use of synthetic biology in agriculture in a safe and sustainable way.
Xenarios, Ioannis
Invited Speaker
Systems biology and network modelling in biological systems: the Vital role of biocuration
Xenarios, Ioannis (1)
(1) Vital-IT, University of Lausanne, CIG, Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland.
During the presentation several activities will be highlighted from the use of biocuration as a mechanism to
structure, contrast and represented knowledge, to the modelling and simulation using logic-based
techniques. I will introduce the way we use predictively such kind of models and how a virtuous cycle
between simulation and experimental validation is put in place. This presentation will also highlight the
need of mixed expertises between bioinformaticians, systems biologists, biocurators and experimentalist
(both clinical and basic researchers).
Zhang, Xiang
Discriminating host defence to fungal and bacterial infections by genome-scale metabolic
modelling of human PBMCs
Zhang, Xiang (1); Mardinoglu, Adil (2); Kuivenhoven, Jan Albert (1); Nielsen, Jens (2); Kumar, Vinod (3);
Wijmenga, Cisca (3); Netea, Mihai (4); Groen, Bert( 1)
(1) University Medical Centre Groningen, Department of Pediatrics, Groningen, The Netherlands; (2) Chalmers
University of Technology, Department of Chemical and Biological Engineering, Goteborg, Sweden; (3) University
Medical Centre Groningen, Department of Genetics, Groningen, The Netherlands; (4) Radboud University Nijmegen
Medical Centre, Department of Medicine, Nijmegen, The Netherlands.
This study aims at discriminating the metabolic pathways induced during induction of host immune
responses by the fungal pathogen Candida albicans or the bacterial stimuli Escherichia coli-derived LPS,
Borrelia burgdorferi, and Mycobacterium tuberculosis (MTB). We have developed PBMC-specific Genomescale Metabolic models (GEMs) for all immune challenges mentioned above. PBMC-specific GEM describes
PBMC’s metabolic physiology with appropriate biochemical, genetic and genomic knowledge of PBMC. In
this study, a PBMC-specific GEM was reconstructed for each immune challenge by applying the tINIT
algorithm [1] based on proteomics and RNA sequencing data of unstimulated PBMCs together with PBMC
microarray data stimulated by Borrelia, LPS, MTB or Candida. Within each metabolic network, the “hot
regions” where significant gene expression changes occurred were identified. Through comparing such “hot
metabolic regions” between fungal and bacterial stimulation, we identified de novo purine synthesis and
page 78 / 88
SB@NL2014 symposium
Symposium booklet
cholesterol biosynthesis as the metabolic signatures of Candida-induced inflammation. We also propose
that squalene 2,3-oxide and 5-aminoimidazole ribonucleotide (AIR), the intermediary metabolites in de
novo purine synthesis and cholesterol biosynthesis can be used as biomarkers for diagnosing a Candidainduced inflammatory response. All the predictions are currently being validated using metabolomics-based
approaches.
[1] Agren, Rasmus, et al. "Identification of anticancer drugs for hepatocellular carcinoma through
personalized genome‐scale metabolic modelling." Molecular Systems Biology 10.3 (2014).
page 79 / 88
SB@NL2014 symposium
Symposium booklet
Partners & Sponsors
Partners & Sponsors
page 80 / 88
SB@NL2014 symposium
Symposium booklet
SB@NL: Netherlands Systems Biology Platform
The aim of the Netherlands Platform for Systems Biology (SB@NL) is to foster the
systems biology approach in the red, green, white and possibly blue sectors of the life
sciences, in particular by creating synergies between systems biology research institutes/groups and other
stakeholders in systems and synthetic biology, biotechnology and medicine.
This platform was built and coordinated by NCSB since 2011, and SB@NL activities are continued independently after
the NCSB programme ended in 2013 (see below, under governance).
SB@NL will work side by side with the bioinformatics community (ex-NBIC) to set-up the Netherlands Bioinformatics
and Systems Biology Research School (BioSB), and to make relevant NCSB training activities (e-courses, tutorials,
workshops) available through the BioSB research school. Other SB@NL activities aim at (a) community development
with strong international connections, (b) communicating with policy makers/granting organisations, (c) event
organisation (e.g. a SB@NL2014 symposium), (d) proposing research programmes/projects/calls, and (e) setting up
the Dutch components of an advanced national and European systems biology research infrastructure (www.isbe.eu).
ZonMw endowed the platform with a modest start-up subsidy (derived from residual NCSB funds).
SB@NL governance
An SB@NL executive committee (‘bestuur’) governs the platform since 2014, taking over the coordinating role
assumed by NCSB since 2011. The executive committee members are Hans Westerhoff (chair), Jaap Molenaar,
Lodewyk Wessels, Edwin de Boer (finance), who are assisted by Diman van Rossum (secretary).
The ‘Council of Partners’ (‘Raad van Partners’), consisting of representatives of the SB@NL partners, will advise
and guide the executive committee. The council convenes roughly once every six months. The first meeting was held
on 20110509 in Utrecht.
The eleven (11) SB@NL partners
CBio – Computational Biology / Biomedical Engineering, Eindhoven University of Technology
City: Eindhoven
Delegates: Natal van Riel, Peter Hilbers
Website: bmi.bmt.tue.nl
Researchers at the department of Biomedical Engineering combine
basic science (how) with engineering (how to). They investigate,
develop and apply engineering principles and tools to unravel the
pathophysiology of diseases and enhance diagnostics, intervention
and treatment.
Interdisciplinary research has been realised in the three thematic
programmes: Regenerative Medicine, Molecular Imaging and Systems
Biology.
The Systems Biology theme aims to enhance the shift from ‘describing’ life’s processes to ‘understanding’ them and
‘capturing’ them in validated predictive models, and even ‘managing’ or ‘controlling’ them. The scope of this theme
page 81 / 88
SB@NL2014 symposium
Symposium booklet
covers all levels from molecules to organs and humans, such as described by the transcriptome, proteome,
metabolome and the physiome.
CMSB – Centre for Medical Systems Biology
City: Leiden
Delegates: Ko Willems van Dijk, Gert-Jan van Ommen
Website: www.cmsb.nl
Common diseases, such as Alzheimer, arthritis, diabetes, migraine and
depression pose major healthcare problems. Such diseases are caused
by a complex network of factors: age, gender and lifestyle play a role in addition to a patient’s genes. The complexity
of common diseases is further increased by the fact that they often interact. At first glance there seems to be no
relationship between depression and migraine, diabetes and cancer, or high cholesterol and Alzheimer. Yet we
increasingly find hidden connections between these diseases, suggesting the existence of common pathways or
biological master switches that underlie multiple clinical outcomes. The main aim of the Centre for Medical Systems
Biology (CMSB) is to elucidate this connectivity. In addition, rare forms of these diseases, often with more clear-cut
genetic causes, are studied as model systems to develop mechanistic insights, better prognostics and targeted
therapies. By combining these approaches, CMSB optimises its mission to improve the diagnosis, treatment and
prevention of common and rare diseases.
CSBB – Centre for Systems Biology and Bioenergetics
City: Nijmegen
Delegates: Richard Notebaart, Gert Vriend
Website: www.csb-bioenergetics.nl
CSBB will model mammalian energy production, distribution,
expenditure and energy stress responses in the context of human
disease.
CSBB focuses its experimental and modelling efforts on the energy
stress response network in mouse and human patient cells, tissues
and at the whole organism level. Multidisciplinairy approaches with
state-of-the-art technologies will be used for modelling of the direct
relationship between processes for ATP production and distribution and ATP expenditure, both under physiological
conditions and following disease-related environmental or genetic pertubations. The developed physiological and
energy stress response models will be used to improve the energy status of man in health and disease via
pharmacological and nutritional interventions.
CSBC – Cancer Systems Biology Centre
City: Amsterdam
Delegate: Lodewyk Wessels
Website: csbc.nki.nl
page 82 / 88
SB@NL2014 symposium
Symposium booklet
From targeted therapeutics to personalised medicine.
The Cancer Systems Biology Center (CSBC) is a center of excellence
within the Netherlands Cancer Institute-Antoni van Leeuwenhoek
hospital. The CSBC will employ computational modeling and
experimental validation cycles spanning multiple levels of complexity
including cell lines, mouse models and patients. Within the CSBC
several research groups from the NKI-AVL work together to develop a strategy to tackle the complexity of molecular
networks that govern breast tumorigenesis with the goal to deliver improved diagnostic tools to enable tailored
cancer therapy.
DCMSB – Delft Centre for Microbial Systems Biology
City: Delft
Delegates: Marcel Reinders, Jack Pronk
Website: dcmsb.tudelft.nl
The Delft Centre for Microbial Systems Biology (DCMSB) harbours
scientists in molecular biology, physiology, enzymology, and bionanoscience, as well as engineers in biotechnology,
bioinformatics and control theory. The DCMSB performs fundamental research on microorganisms and microbial
populations, inspired by current and future industrial applications. Over the past decade, DCMSB participants have
fruitfully co-operated in quantitative modelling of (populations of) microorganisms within (a) the Delft Research
Centre for Life Science&Technology, (b) substantial national research programmes such as B-Basic, and (c) the Kluyver
Centre for Genomics of Industrial Fermentation. Involvement in these large programmes stimulates efficient
dissemination and valorisation of DCMSB research. Note that the DCMSB is complementary to these efforts, as it
uniquely offers its members an environment dedicated to the study of basic biological phenomena unencumbered by
commercial or competitive considerations. Without the DCMSB, this research cannot be performed at the scale and
level of collaboration between experts in biology, measurement technology and modelling we believe is required to
obtain success.
MaCSBio – Maastricht Centre for Systems Biology
City: Maastricht
Delegate: Ilja Arts
Website: macsbio.maastrichtuniversity.nl
Maastricht University and Maastricht University Medical Centre have recently decided
to strengthen their systems biology research. With support of the Province of Limburg,
the Maastricht Centre for Systems Biology (MaCSBio) was founded. The centre will
combine modelling expertise and biological research including behavioural research
from the Faculty of Health Medicine and Life Sciences, the Faculty of Psychology and Neurosciences, and the Faculty
of Humanities and Sciences. The primary aim of MaCSBio is to facilitate the integration of relevant biological data
coming from different empirical domains using mathematical multi-scale modelling approaches. This will take shape in
two research lines, which will be established within the centre. The focus of the research lines will be determined at
the end of 2013 and research will start in 2014.
page 83 / 88
SB@NL2014 symposium
Symposium booklet
NISB – Netherlands Institute for Systems Biology
City: Amsterdam
Delegates: Bas Teusink, Roeland Merks, Age Smilde
Website: www.sysbio.nl
Within the Netherlands Institute for Systems Biology (NISB),
researchers from the University of Amsterdam, VU University
Amsterdam and Centrum Wiskunde&Informatica collaborate to develop and exploit systems biology tools to
investigate complex biological systems and unravel their underlying principles. NISB researchers approach biological
systems as networks of molecules, cells, tissues and organisms that interact in time and space. This is achieved by
amalgamating biological and biomedical sciences with chemistry, physics, mathematics, informatics and engineering.
SBC-EMA – Systems Biology Centre of Energy Metabolism and Ageing
City: Amsterdam
Delegates: Matthias Heinemann, Barbara Bakker
Website: www.rug.nl/research/centre-for-systems-biology
The goal of SBC-EMA is to unravel the intricate relationship between
metabolism and the biology of ageing. This relationship, which has
been conserved throughout evolution, is bidirectional: metabolism
affects the ageing of an organism, while the metabolic system itself
declines under the influence of ageing. Both metabolism and ageing
are systems properties that are affected by a large number of processes at the molecular, cellular and organ level.
Hence a full understanding of metabolism, ageing and their interaction necessitates a systems-biology approach.
We use the budding yeast Saccharomyces cerevisiae and the mouse as model organisms. With an interdisciplinary
team of scientists,we will use various approaches of systems biology.
MSB-TNO – Microbiology & Systems Biology (TNO)
City: Zeist
Delegate: Suzan Wopereis
Website: www.tno.nl/msb
Microbiology & Systems Biology (MSB) is internationally acclaimed for
finding innovative solutions for diverse challenges in nutrition, health
and biotechnology. MSB uses its broadly applicable expertise to create tailor-made innovations for companies in
agro&food, personal care, chemicals, biotechnology and pharma. We use systems biology to understand relationships
in complex biological processes. We are particularly successful in developing indicators of biological and pathogenic
processes (inflammation, food spoilage). In collaboration with other TNO groups we have access to a broad collection
of state of the art analytical platforms which we apply to an extensive collection of in vitro and in vivo models of
health and disease. We use this to solve complex problems for industrial partners. For example, we determine
whether food ingredients have healthy effects on gut and mouth microbiota or have health-promoting activity in
animals or humans or mouse models for cardiovascular diseases. We also use systems biology to find natural food
ingredients that can suppress food spoilage or prevent food poisoning. In addition, we use systems biology to optimise
page 84 / 88
SB@NL2014 symposium
Symposium booklet
industrial bioprocesses and develop methods to convert biomass into commodity molecules (building blocks and
biofuels).
WCSB – Wageningen Centre for Systems Biology
City: Wageningen
Delegates: Vitor Martins dos Santos, Jaap Molenaar
Website: www.wageningenur.nl/systemsbiology
Wageningen UR has chosen Systems Biology as one of its focal points
and invests considerably in this emerging research field. The
Wageningen Centre for Systems Biology (WCSB) for Food, Feed, and
Health is the condensation point of all Systems Biology developments
at Wageningen UR.
WCSB’s research focuses on the entire spectrum of biological systems, from DNA to ecosystem, and involves topics
relevant to plants, digestion and microbes.
With the establishment of the WCSB and the financing of 14 systems biology research projects under its auspices,
Wageningen UR aims to become a strategic partner in the field of research in systems biology. The projects fall apart
into three categories:
a. Learning and predicting how plants respond to stress and the consequences this has for yield and productivity
(Virtual Plant)
b. Understanding the functioning of the intestinal tract of mammals in relation to their diet to anticipate or remedy
obesity (Virtual Gut)
c.Developing models that can help in the modification of microorganisms for the production of fine chemicals,
commodity compounds and energy (Virtual Microbe).
SMEs – Small and Medium Enterprises
SB@NL recognises the importance of valorisation of systems biology knowledge and its effective translation into
solutions which benefit the society and economy. SMEs play a crucial role in valorisation being the key drivers of
innovation. Hence, SB@NL involves the SME-world in its partnership through representative(s) of SMEs that are active
in the systems biology field.
City: Utrecht
Delegate: Marijana Radonjic (EdgeLeap BV)
Website: www.edgeleap.com
EdgeLeap B.V. is an Utrecht-based company providing cutting edge
bioinformatics solutions for the life sciences industry. EdgeLeap
applies network-based technology for data integration, mining and
visualisation to uncover how multiple layers of biological complexity
relate, influence each other, and organise themselves into complex
systems. This knowledge helps making decisions based on bestevidence for designing strategies to cure disease or improve health. EdgeLeap reinforces the competitive edge of its
clients through extracting actionable knowledge out of their complex data.
Founders of EdgeLeap are scientists with years-long involvement in national systems biology platforms (e.g. NgC,
page 85 / 88
SB@NL2014 symposium
Symposium booklet
NBIC, NTC, NGI). The startup of EdgeLeap has been boosted by the Venture Challenge programme of the Netherlands
Genomics Initiative (NGI), fostering the combination of excellent science with entrepreneurship
(www.lifesciencesatwork.nl). In parallel with running a business, the EdgeLeap team remains to be active in the
academic network through co-supervision of PhD students, editing scientific journals, co-organisation of symposia and
workshops, providing training and education, and, acting as co-developers in open source communities.
page 86 / 88
SB@NL2014 symposium
Symposium booklet
Sponsors
This symposium is endowed by the generous financial support of the following organisations.
1. ERA-Net for Applied Systems Biology (ERASysAPP)
2. Netherlands Systems Biology Platform (SB@NL)
3. Netherlands Institute for Systems Biology (NISB)
4. Netherlands Consortium for Systems Biology (NCSB)
5. Maastricht Centre for Systems Biology (MaCSBio)
6. Maastricht University (UM)
page 87 / 88
SB@NL2014 symposium
Symposium booklet
page 88 / 88

SB@NL2014 abstract book (version dd

Transcription

Similar documents

Please contact our Global Event Associate, Thu Thao Nguyen: E

FERNANDEZ STOHANZL PORTFOLIO 2014.indd

the netherlands

Opera Graphic Design Tempo

Mycology Update - Vereniging voor infectieziekten

10 species of sea anemones

Stadsschouwburg, Amsterdam - Rabo Zaal 2009

Emergency Radiology Symposium - Continuing Medical Education

Bouldering at Amsterdam

Linguistic Data Cleansing Case Studies