Self-evaluation Research Programme Bioinformatics


1 Οκτ 2013 (πριν από 4 χρόνια και 7 μήνες)

156 εμφανίσεις


Programme Bioinformatics


Programme: Bioinformatics

B. Documentation regarding the level of the research programme

Research area and mission
: Performing high
quality and
internationally visible research i

Bioinformatics and Systems Bioinformatics, including a clear focus on developing
bioinformatics tools.

ACM Code:

J.3 (Life and Medical Sciences)

Programme leaders:

Prof. dr. J. Heringa, Prof. dr. B. Teusink

Starting date of the programme
: 01/11/
2002 (FEW
Computer Science
Section), 01/01/2006 (IBIVU Centre)

Affiliations outside the Institute

Heringa is at
VU since October 2002 and is appointed for 0.8 fte at the Computer Science
department of the Faculty of Exact Sciences (
FEW) and for 0.2 fte at the Department for
Molecular Cell Biology (MCB) within the Faculty of Earth and Life Sciences (FALW).

Other affiliations:

Netherlands Bioinformatics Center (NBIC): PI, Subprogramme leader, Member of
NBIC Management Team, Chairman

of the NBIC Education Committee

Centre for Medical Systems Biology (CMSB): PI

Ecogenomics, PI, Workpackage leader

Netherlands Institute for Systems Biology (NISB): PI

EC Framework Project 6 Network of Excellence ENFIN: PI

Teusink is at the VU since Augus
t 2008 and is appointed for 0.2 fte at the Computer Science
department and for 0.8 fte at the Department of Molecular Cell Biology (MCB) within

Other affiliations:

Netherlands Bioinformatics Center (NBIC): PI

Kluyver Centre for Genomics of Industr
ial Fermentation (KC): PI and programme

Ecogenomics, PI

Netherlands Institute for Systems Biology (NISB): PI, Member of Management Team

Netherlands Consortium for Systems Biology (NCSB): PI, chairman of the Internal
Advisory Board



This self
evaluation will focus on the Bioinformatics research activities conducted in or
relevant for the Bioinformatics section within the Department of Computer Science.
However, the Bioinformatics section is an integral, intimately connected and esse
component of the IBIVU centre. This means that insight in significance of the research
endeavour can only be obtained by means of a comprehensive exposure of the IBIVU
activities. In the remainder of this self
evaluation, it will be indicated to what

extent a given
project or activity is embedded within the Computer Science department.

Bioinformatics is a very wide and versatile area of research, with a number of neighbouring
disciplines. One of the fundamental challenges in bioinformatics is bridgin
g the gap between


these disciplines, particularly between experimentalists on the one hand, and theoreticians and
computer scientists on the other hand. In Section B.2 below, the structure of IBIVU is
described, which embodies the organisation structure of

the Bioinformatics research
conducted at the Vrije Universiteit. The IBIVU has been founded by the Faculty of Sciences
(FEW) and the Faculty of Earth and Life Sciences (FALW). The IBIVU centre ties together a
number of research lines that are partly in in
itiation and partly in the planning stage. The
bioinformatics research is organised in a number of research lines or
scientific themes
, each
headed by a chair person. The IBIVU centre’s director is Prof. Heringa, who was involved in
setting up the IBIVU f
rom its inception. This includes ongoing research planning discussions
with various parties within the Vrije Universiteit (including the VUmc) and (inter)national
groups, interdisciplinary grant acquisition, organizing the research infrastructure, and dail
management of the IBIVU. Since September 2003, a 2
year Integrative Bioinformatics master
programme is in operation. Recently, the IBIVU also started the operation of a 2
interdisciplinary MSc programme in Systems Biology. The IBIVU centre has rece
ntly been
extended by a second group in Systems Bioinformatics, led by Prof. Teusink. To ensure tight
collaboration and cross
fertilisation between the founding faculties, Heringa is appointed for
0.8fte in FEW and 0.2fte in FALW, while Teusink acts recipr
ocally with 0.2fte embedding in
FEW and 0.8fte in FALW.

The projected inclusive size of the IBIVU (resulting from internal/external funding) and the
many national and international collaborations make it vitally important to set up the right
structure and to have extensive and effective communication among the IBIVU
centre’s members.

Our management style aims to create a team
spirit where people can benefit as much as
possible from each other's work and can make fruitful contributions to the

group as a whole.
This is of vital importance given the widely divergent expertise present within the IBIVU
staff. Every Wednesday morning, we have a group meeting where we discuss research
progress and challenges, while fortnightly an (informal) in

research discussion takes
place and on alternating weeks a literature discussion is being held, the latter including
scanning of significant papers and in
depth discussion of selected papers.

Leadership is ultimately about creating a way for people to c
ontribute to making something
extraordinary happen. Within Bioinformatics research, which is inherently interdisciplinary, a
challenge with staff entering the field from a neighbouring discipline is the breadth of the
research area, such that sometimes the
y may experience difficulty in knowing how the parts
fit in the whole of their work. It is therefore vitally important, for example, to show junior
group members how the components in bioinformatics research are structured, such that they
develop a sense o
f the integrated bioinformatics landscape with all its related disciplines (e.g.
molecular biology, computer science, mathematics (statistics), physics, etc.). This means that
an informal style of leading the group is called for, with a good sense of atte
ntion for
individual needs of group members. Group members are actively endorsed to go on courses
and (inter)national conferences or workshops in order to increase their expertise and widen
their scope.

Within the Teusink group multidisciplinarity h
as an even larger scope, as this group also
performs wet experiments to collect data sets. This requires very careful balancing of
expertises and characters within the research team, from PhD level to staff. Also effective
communication between the group m
embers is key. All team members are sent to courses


to make sure they can at least understand the language of the different disciplines
involved. Within the possibilities of the building and space, people are housed strategically
together, also

mixing staff from the Heringa and Teusink groups, to maximize interaction.
Besides the Wednesday group meetings, there are specific topic
focussed meetings with
experimentalists, modelers and bioinformaticians together. Central in the leadership of the
oup is to facilitate the talent to flourish and create excellent, multidisciplinary research.



Strategy and policy

In 2001, FEW drafted an initial plan (co
ordinated by prof. Bal) and applied for funding with
the VU board to set up a Bioinformatics U
nit within FEW. The VU board welcomed the
proposal, but advised to widen the proposal and include FALW. In 2002, the Boards of FEW
and FALW decided to prepare an initiative to found a new broad based institution for research
and teaching in Bioinformatics
. Based on a business plan by Profs. Jaap Heringa and Hans
Westerhoff, the joint Faculty Boards decided towards the end of 2002 to submit the plan to
the Board of the University (CvB) for approval and funding. Subsequently, preparations for
founding the C
entre for Integrative Bioinformatics (IBIVU) as an interdisciplinary centre for
research and education in Bioinformatics were initiated in 2003 by FEW and FALW. Both
faculties support the IBIVU financially with about 1.4 M€ over 5 years. In September 2005,

the CvB approved the IBIVU Implementation Plan and endorsed the IBIVU initiative with
another 1.5 M€ over 5 years, while appointing FALW as the managing faculty (‘Penvoerder’).
Following this success and subsequent preparations to start the IBIVU and recr
uit the second
chair, the official ‘average’ starting date for the IBIVU has been set by the IBIVU Governing
Board and founding faculties to January 1
, 2006.

The last few years have witnessed two important national developments that bode well with
emergence of the IBIVU Centre. In 2006, the Netherlands Bioinformatics Centre (NBIC)
has been founded, based upon funding by the Netherlands Genomics Initiative (NGI). NBIC
underpins and brings together national bioinformatics initiatives. It comprises thr
ee pillars:
BioRange (bioinformatics research), BioWise (bioinformatics education) and BioAssist
(bioinformatics infrastructure and support). Heringa is a member of NBIC’s mangement team.
Another crucial development is the start of the Netherlands Institut
e for Systems Biology
(NISB), founded by the four Amsterdam institutions VU, UvA, AMOLF and CWI. An
obvious implication of this is that the NISB is actively endorsed by the VU board. During its
inception, Heringa has been in the NISB co
ordination committe
e, a role that has recently
been taken over by Teusink as member of the NISB Management Team. The IBIVU plays a
leading role in the NISB as it hosts the NISB senior chair (Teusink), while also the NISB
junior chair (Bruggeman) is partly embedded in the Teu
sink group. The NBIC and NISB
initiatives have led to research funding awarded to both the Heringa and Teusink groups,
while they also represent ideal platforms for collaboration and streamlining integrative and
systems bioinformatics research.

The b
based setup of the IBIVU has not only allowed the seamless integration of the
NBIC and NISB activities, but also provides an essential platform for Bioinformatics
initiatives at the VU. Within the IBIVU, a number of exciting research lines come togeth
and a number of staff from the various collaborative research projects are drawn into
Integrative Bioinformatics and Systems Bioinformatics research. An example of this is the
fact that two PhD students (Binsl and Hettling) and a postdoc (Gavai) awarde
d to Dr. Hans
van Beek (VUmc) are embedded and housed within the IBIVU centre in order to reap the
benefits of the IBIVU synergy.


IBIVU mission

The IBIVU mission is to establish and maintain an excellent Integrative Bioinformatics
research programm
e at the VU. This implies scientific research and teaching in Integrative
Bioinformatics and Systems Bioinformatics, both of international excellence. This should also
lead to the coordination of large
scale applications for research funding at the nationa
European and international level.


Strength and positioning of the IBIVU

The IBIVU organisation provides a competitive edge in Bioinformatics in The Netherlands in
three important ways:


Its inclusive organisation structure is unique and appropriate for

carrying out true
integrative bioinformatics research
which is inherently multi

at three


Integrating the various biomedical and genomics data sets with each other
and with biophysical and biochemical principles in order to (i) level

that impede high
level understanding of cellular processes and (ii) advance
drug design, treatment of disease, and biotechnology


Integrating new bioinformatics tool creation with advanced experimentation
in order to facilitate large
scale integra
tive genomic analysis modes


Integrating teams with various expertise from and throughout the
participating faculties


All bioinformatics activities in the Netherlands and the vast majority worldwide are
embedded in biomolecular and/or biomedical department
s. Few integral
bioinformatics units are part of a Computer Science or Informatics department. The
IBIVU has an important component embedded in the Computer Science department
at the Faculty of Sciences, which creates significant opportunities with regard

advanced tool creation, but even more uniquely, this component is integrated with an
equal size component in a biomolecular department. The joint positioning of both
IBIVU professors (see B1.1) also attest to this fact.


Over the last few years, Systems

Biology has gained a lot of momentum. Although
modelling of complex dynamical processes in biological systems goes back decades,
the last few years have seen an awareness that the classical modelling approach based
upon ordinary differential equations (OD
Es) can and needs to be extended. The
IBIVU positions itself at the interface of bioinformatics and systems biology, which
we coin Systems Bioinformatics, developing new computational infrastructure and
approaches of significance for the research area. An
additional focus is constraint
based modelling, where constraints abounding from physics, chemistry and biology
are incorporated to make the modelling parameter space tractable. Constraint
modelling is a new focus of the IBIVU research that has been
initiated following the
arrival of the Teusink group. Such approaches are being taken at various levels, from
scale networks to genome
scale metabolic networks.

At the highest level, a primary scientific objective of the IBIVU centre is understandin
g the
modes of action of the integrated levels of cellular networks (e.g. regulation, miRNA,
signalling, metabolome) leading to emergent properties in the cell. For example, it is
becoming increasingly clear that many complex cellular networks show bistabl
e switching
behaviour, or even multi
stable switches, which can explain many of apparantly paradoxical
characteristics of the networks, such as time delays in bacterial growth after environmental
changes, or differentation of various types of stem cells.

The integrative and tools
directed positioning of the IBIVU outlined above provides a strong
profile for the international 2
year master programmes in Bioinformatics and in Systems
Biology, and teaching activities at the bachelor level. As Bioinformatics
is a quickly
developing research area, the broad
based set
up of the IBIVU creates an ideal platform for
continuously updating the Bioinformatics and Systems Biology master programmes and for
student internships in various new and challenging research area


Mission of the bioinformatics section

The mission of the bioinformatics section is to develop state

art bioinformatics
prediction, data mining and modelling methods that allow us to build up fundamental
understanding of biological syste
ms. These methods will be compliant with execution on


performance distributed systems, such as cluster computers, grids, and networks. Our
research strategy is to

conduct fundamental research into appropriate (systems) bioinformatics areas and
integrative tools

develop realistic applications for our software by collaborating with biological
application groups

guarantee the impact of our work not only by publishing papers, but also by making
servers (see and softw
are distributions that are used by
other researchers.

have strong participation in international research communities

test and evaluate new biological insights and models experimentally. The Teusink
group has acquired an experimental infrastructure that al
lows experimental
measurement of variables (gene expression, metabolites, enzyme activities). This is
crucial for the development and validation of genome
wide metabolic models of
various microorganisms and
in the near future

microbial communities.

Some policies we follow to implement this strategy include:

we aim to publish in top international journals

we explicitly do not aim to produce large numbers of (incremental) papers and focus
on quality, not quantity

we give a high priority to building res
earch infrastructures such as web servers.

Integrating Genomics, Bioinformatics and Systems Biology

A key ingredient for building up understanding of cellular behaviour is integrating genomics,
bioinformatics and computational systems biology (C
SB). The main challenge is the
integration of component
based high
throughput genome
wide data pipelines and associated
bioinformatics analysis modes (top
down approach) on the one hand, with systemic modeling
approaches at the cellular networks level (bot
up approach) on the other. This ‘middle
approach’ will require the development of new combined data
mining and modeling
formalisms. The IBIVU centre is well positioned to tackle this issue: in addition to its core of
oriented researchers, it

takes part in various international networks that engage in
computational systems biology and bioinformatics. These include the EU Networks of
Excellence “ENFIN” (, which focuses on “enabling systems biology”, and
‘BioSim’ focusing on the si
mulation in systems biology of drug discovery, the EU Marie
Curie training networks on Nuclear Receptor Systems Biology (‘Nuc Sys), the Yeast Systems
Biology Network (An EU Concerted Action) which has been extended in the FP7 UniCellSys
Network. These netw
orks include bioinformaticians, computational systems biologists, as
well as experimental groups, with as a main objective to foster the model building /
experimental validation cycle with a comprehensive bioinformatics and systems biology
Moreover, the Section of Molecular Cell Physiology (MCP) of the Department
for Molecular Cell Biology (MCB) at FALW has a strong focus on Systems Biology and
hosts computational as well as experimental scientists, such that this combined set
up offers a
ngible opportunity for integrating Bioinformatics and Systems Biology. Both Teusink and
Heringa are therefore member of this section for the FALW part of their appointment.
Furthermore, the Department of Computer Science and the Department of Mathematics a
FEW include key research sections that participate in the IBIVU and can be instrumental in
developing core mathematical and computational tools (e.g. Section Computer Systems,
Section Artificial Intelligence, Section Theoretical Computer Sciences, Sectio
n Stochastics).

The integration of Genomics, Bioinformatics and Systems Biology aims at elucidating how
the functional properties of living organisms arise in the interactions of their genome
components. Since the interactions are nonlinear and pot
entially genome wide, this novel
science requires a more systematic approach than has been common to biology. This model


building/experimental validation cycle is also referred to as the "four M's" of Systems
Biology: measurement, mining, modelling, and m
anipulation. Manipulation and measurement
are on the experimental side, whilst mining and modeling are on the computational side. The
four M's are part of an iterative cycle, beginning with manipulating the system. Once a system
is perturbed, it is measure
d using a high
throughput, multivariate technology. The data are
then mined (in conjunction with existing genomics and systemic data) to elucidate hypotheses
that, when cast in terms of formal computational models, can form the basis for a new round
of sys
tem manipulation. Once an appropriate model has been developed, the outcome of
manipulating the system can be predicted. Successful iterations of the above kind then lead to
the formation of new understanding of, and theories on, living systems. The theor
ies may then
again be tested and applied by iterative cycles.


Processes in research, internal and external collaboration

One of the challenges of the IBIVU as a young research centre is the relatively large number
of different research projects laid

down in the IBIVU Implementation Plan. The cross
structure of the IBIVU and its organization of research themes is described in Appendix 1
below. A number of research themes are in the start
up phase, for example within the FALW
component of the
IBIVU or other institutions at VU. All of these projects draw attention and
help from the core Bioinformatics Section at the Department of Computer Science. In light of
this diversity it is no easy task to create an environment where researchers can collab
intensively and benefit optimally from each other's work. We nonetheless attempt to
accomplish this by processes explained for both IBIVU groups in the following sections.


Processes in research

B3.1.1: Heringa group

An important instrument
for internal collaboration is to have a common
hardware/software infrastructure, much like experimental scientists collaborating in a
common lab.


Group members develop software generally under Linux, that is made to run
on the IBIVU cluster computer (IBIS


We try to keep the number of different computer languages manageable.


We have an extensive IBIVU website ( where the software
produced by the group is made available via user
friendly web servers and
downloadable codes.

We focus on a num
ber of central areas within Bioinformatics, e.g. multiple sequence
alignment, protein structure prediction and integrative algorithms deriving from the
former two areas. A key advantage is that most if not all collaborative projects
depend on these areas.
Moreover, the focus has resulted in an international reputation
in areas such as:


Multiple sequence alignment (PRALINE and T
Coffee methods)


Secondary structure prediction (SSPRED and YASPIN methods)


Sequence repeats detection (REPRO, TRUST and RALIGN meth


Protein functional specificity site prediction (SH and MultiRELIEF methods


Protein domain boundary prediction (SnapDRAGON, Domaination and
DOmain methods)

A relatively recent focus is on generating a computational infrastructure for formal
modelling of (multi)cellular processes using Petri nets. In collaboration with the Bal
group (Computer Systems) and the Fokkink group (Theoretical Computer Science),


we are developing new formalisms and an integrated infrastructure (including
dynamic visua
lisation) for Petri net modelling.

In addition to these instruments, we have several types of meetings to stimulate collaboration,
ranging from individual meetings to a weekly group meeting (so all group members know
what their fellow members are doing)

and at least a weekly joint lunch following the group
meeting. Ph.D. students have a meeting with their supervisor at least every week. For most
Ph.D. students, we use a postdoc as extra supervisor, who also attends these meetings. Quality
improvement i
n research includes dissemination of results: the supervisors always actively
participate in publications of beginning Ph.D. students and give detailed feedback on the
methodology as well as the (textual or oral) presentation. In addition, we have colloq
uia and
frequently scheduled project meetings as well as more informal brainstorming meetings for
Ph.D. students.

The IBIVU Implementation Plan details project
wise associations with the following VU

MCB (Department of Molecular Cell Biolo

Systems Biology

CNCR (Centre for Neurogenomics and Cognitive Research)

CRBCS (Centre for Research on BioComplex Systems)

VUmc (VU Medical Centre)

In addition to the direct funding by FEW, FALW and CvB of the VU, the IBIVU receives
funding by taki
ng part in the following (inter)national initiatives:

CMSB (Centre for Medical Systems Biology), DIAL (Data Integration, Analysis and
Logistics) project

Ecogenomics project

NBIC (Netherlands Bioinformatics Centre), BioRange project


SP 2.2.1


SP 2.3.1


SP 3.



EC FP6 Network of Excellence ENFIN (Experimental Network for Functional

B3.1.2 Teusink group

The Teusink group started in August 2008 and is positioning itself as systems bioinformatics,
i.e. systems biology with a special focus on

integrative bioinformatics. It aims at forming
bridges between the classical bottom
up (ODE
based) approaches in computational biology
and the more data
driven approaches in classical bioinformatics. Within the group, focus is
created by:

Research questio
ns: we focus on the use of constraint optimization techniques to
investigate what aspects of biological systems have been optimized through evolution
to the extent that physical and/or chemical boundaries are hit (which evolution
obviously cannot change).
Optimality is potentially very powerful for reducing
solution spaces, and for understanding design principles of biological systems. On the
experimental side, we have focused on long
term cultivation infrastructure to allow
selection of cells that have opt
imally adapted to specific environments imposed on

Tools and models: we work at different levels of detail and abstraction, but with the
specific goal to integrate these approaches.


scale metabolic reconstructions and models are used for the
picture and integration of high
throughput data



called self
replicating models: these are modular models able to capture
offs in whole cells that govern growth strategies in unicellular


detailed kinetic models of subsystems


dge base of biologically relevant physico
chemical constraints
(general for each cell) and more specific constraints for the model organisms

Within our group we focus on a limited number of model organisms: these include
Lactococcus lactis
Saccharomyces c

and to some extent
Escherichia coli
The tools that we develop while studying these organisms are then applicable to other
organisms as well (e.g. AUTOGRAPH, our method to accelerate the construction of
scale metabolic models
PMID: 1677202

In addition to the interactions and funding described above, the Teusink group receives
funding by taking part in the following (inter)national initiatives:

Kluyver Centre for Genomics of Industrial Fermentation

NBIC (Netherlands Bioinformatics Centr
e), BioRange project SP3.7.1


Computational Life Science grant

2 STW projects, one on optimization of
L. lactis

as cell factory, another on population
heterogeneity in
L. lactis

EC FP7 Network of Excellence UniCellSys

LAB Systems Biology of Micr
oOrganisms project


Internal collaboration

The Bioinformatics section has several ongoing or planned collaborations inside the
Department of Computer Science and FEW.

B3.2.1 Heringa group

Collaboration with the Computer Systems Group (Prof. Bal) re
sulted in new ways of
parallelizing Bioinformatics software and implementing it on a cluster computer.
More details are given in Section B.9. We are also sharing students from the top
master PDCS, on implementing bioinformatics pipelines on GRIDS and clust
computers. Projects also include the development of a bioinformatics work flow on a
cluster computer.

Further collaborations are with the Artificial Intelligence group of Prof. Treur.
Projects include agent
based modeling of dynamically connected system
s in
ecogenomics and neurobiology.

We recently agreed on starting a collaboration with the group of Prof. Van Harmelen
(Artificial Intelligence) on ontologies in biodata. A postdoc to be hired on the ENFIN
EC FP6 NoE will also work in this area.

tion with the Department of Chemistry (Prof. Vermeulen) is set up within
the framework of the Top Institute Pharma (TIPharma) national research initiative
and that of the new Amsterdam Institute for Molecules, Medicine and Systems

Collaboration wi
th the Department of Mathematics is directed at data mining for
proteomics in cooperation with the new VUmc Cancer Center Amsterdam (CCA).
See Section B.6 for more detail.

B3.2.2 Teusink group

The Teusink group started only recently. So far, collaboratio
ns have started with Prof.
Hubertus Irth (Bioanalytical Chemistry), Prof. Martine Smit (Pharmacochemistry), and Prof.
Leen Stougie (Economy, Center for Mathematics and Computer Science, CWI).


Collaborations within the Department of Computer Science are pla
nned, as well as
collaborations with the Laser Center (biophysics, structural biology). Teusink was part of the
management team that drafted the blueprint for the new
Amsterdam Institute for Molecules,
Medicine and Systems (AIMMS).

B3.3 External collabor

The Bioinformatics section has a number of collaborations outside FEW.


Heringa group

We collaborate on Systems Biology together with the FALW Department of
Molecular Cell Biology (MCB). This includes joint staff between IBIVU and MCB.

We c
ollaborate with the VU/VUmc Center for Neurogenomics and Cognitive
Research (CNCR) on brain network modeling. This has resulted in two joint UDs
(Dr. van Ooyen and Dr. van Nierop), between CNCR and IBIVU (FALW matching
and CvB subsidy to FALW). The van O
oyen group is currently expanding as a result
of research funding (e.g. NWO).

We participate in the national Ecogenomics initiative, which has resulted in IBIVU
funding embedded at FALW for a Ph.D. position (van Houte), PostDoc position
Kaplon) and

a postdoctoral position (Tas).

Collaboration with VUmc has been set up to develop a bioinformatics pipeline for
CGH data.

We are discussing intensified collaboration (including joint faculty) with the new
VUmc Cancer Center Amsterdam (CCA) (Dr. Ji
menez) on proteomics research. We
are developing data mining techniques for mass spectrometry data. The collaboration
with the group of Dr. Jimenez resulted in a NWO
Horizon grant (Marchiori).

We collaborate with 18 European teams at the forefront of Bioi
Biology research within the framework of the EC FP6 NoE “ENFIN”. This has
resulted in funding for a postdoc for 3 years, which recently became extended by
funding for an additional 2.5 years. Current topics include co
developing the ENFI
computational infrastructure, protein domain detection, phosphorilation
protein interaction prediction, and Fanconi Anemia gene prediction and
network reconstruction.

An additional collaboration extending out of the ENFIN network is with
Dr. Bertie
Göttgens (Cambridge Cancer Centre, University of Cambridge, UK) and the group of
I. Xenarios (Swiss Institute of Bioinformatics, Lausanne) on modelling early cell
fates in the development of
haematopoietic stem cells (blood stem cells). Dr.
tgens is an international expert on
Haematopoiesis. We aim to combine
Boolean modelling techniques with our novel bounded Petri
net approach (see
below) to unravel and reconstruct the gene network leading to blood cell
differentiation, where 50 known diffe
rent stable gene activity states should be

Together with the group of Prof. Frishman (TU Munich), we collaborate on a
classification system and database of transmembrane proteins.

B3.3.2 Teusink group

Within the BioRange program (Netherlands
Bioinformatics Centre), collaboration
exists in project SP
3.7.1 on integrative bioinformatics. This project focuses on
handling, integration and analysis of omics data. Partners are RUG (B. Poolman, O.
Kuipers, J. Kok, J. Roerdink), UvA (A. Smilde, T. Bre
it), and RU Nijmegen (R.


Siezen), 1 PhD (Notebaart, graduation at RUN). In the second BioRange round, we
collaborate with a.o. Smilde (UvA), 1 postdoc, vacancy.

On lactic acid bacteria collaborations exist with P. Hols (Louvain la Neuve), J. Snoep
bosch, South Africa) and P. Jensen (Copenhagen). In the Netherlands, there
is a STW project together with Hugenholtz (Univ. of Amsterdam), Kuipers and
Poolman (Groningen) and De Vos (Wageningen), resulting in 1 PhD (Goel, guest).

Together with O. Kuipers (
RUG) we have recently acquired STW funding for single
cell analysis of
L. lactis

in relation to population heterogeneity; 3 year postdoc

Within the European Systems Biology of Microorganisms program (SysMo) Teusink
collaborates with the groups
of J. Hugenholtz (Univ. of Amsterdam), W. de Vos
(Wageningen), U. Kummer (Heidelberg), H. Westerhoff (Manchester), I. Nes (Aas)
and B. Kreikemeyer (Rostock) on comparative systems biology of three lactic acid
bacteria; 3 years postdoc (Musters, location Wa
geningen). Teusink coordinates the
draft of a follow
up proposal from this consortium in round 2 of SysMo.

Within FP7 we collaborate with 16 European teams on systems biology of yeast
(UniCellSys); 5 years postdoc (van Grinsven).

We participate strongly in

the Kluyver Center, resulting in 2 PhD positions (Ozturk,
vacancy) and 1 postdoc (Santos, as visitor).

Collaborations exist with Delft (group Reinders) within an NWO
CLS project (only 5
out of 40 proposals awarded).


Academic reputation

B4.1 Hering
a group

Heringa is an editorial board member of the following international peer
reviewed journals,
among which is
, currently the most prestigious journal in bioinformatics:


Bioinformatics and Biology Insights

Computational B
iology and Chemistry

Current Peptide and Protein Research

Evolutionary Biology Online

International Journal of Data Mining and Bioinformatics

Heringa is also in the Editorial Advisory Board of Wiley’s Encyclopedia of Life Sciences.

Since 1998, Heringa ha
s served as expert evaluator of EC Framework Project 5 and
Framework Project 6 proposals. Additionally, he has been an external evaluator on a number
of EC FP5 project mid
term reviews. He has also served on the board of the USA National
Institute for Heal
th (NIH) as an expert evaluator of NIH
NIAID Resource Center proposals.
Since December 2005 he acts as an expert evaluator of the Deutsche Forschungs
Gemeinschaft (DFG) Excellence Initiative. He serves as a member of the national
NWO/Horizon and VENI commi
ttees. He has also acted as an external specialist in a
Visitation Committee (co
ordinated by VBI Hobeon) for the accreditation of the
Bioinformatics BSc programme of the
University of Applied Sciences Leiden (Hogeschool

Heringa serves and has s
erved on longer
term advisory boards, including his participation as
an external consultant for setting up a Bioinformatics Centre, Hamburg University, Germany,
and as an Advisory Board member for the research programme
Bioinformatics for the
Functional An
alysis of Mammalian Genomes (BFAM),

Germany. Currently, he is member of


the Management Team of the Netherlands Bioinformatics Center (NBIC), for which he also
serves as Chair of the Education Committee. Since 2006, Heringa is a member of the
Education Adv
isory Committee (Opleidings Advies Commissie) for bioinformatics of the
University of Applied Sciences Leiden (Hogeschool Leiden). In November 2007, within the
framework of NBIC and under auspices of the Dutch Embassy in Tokyo, Heringa was
member of a dele
gation that visited a number of institutions and governmental agencies
throughout Japan to discuss possible collaborations in bioinformatics between Japan and the

Furthermore, Heringa regularly participates in Programme Committees of intern
workshops and conferences, most notably the annual and largest bioinformatics conference
Intelligent Systems in Molecular Biology (ISMB), as well as the European Conference on
Computational Biology (ECCB). Since a number of years he is one of the c
hairs for the ISMB
Highlight Track
. He has also been a Governing Board member of the national NVBMB
Working Group for Bioinformatics.

He has been a keynote speaker at the
3rd International
Symposium on Computational Life Science
, Utrecht (2007) and at the
Virtual Discovery
Europe Conference, Amsterdam (2008).

(UD from 01/11/2002


Invited lectures at various European and Japanese universities and research institutes,
as well as international conferences.

Member of the examination co
mmittee of PhD defenses at various universities

PI or co
investigator on international research grants (e.g. BBSRC).

Invited guest researcher for the Japanese Foundation of Science.

(UD from 15/05/2006


of the annual
Bioinformatics/Data Mining workshop

Invited lectures at various European universities and research institutes, as well as
international conferences.

PI or co
investigator on national (e.g. NWO
Horizon) and international research

(e.g. EC FP5 SURPRISE network).


(UD as of 01/01/2007)

Invited lectures at various European universities and research institutes, as well as
international conferences.

PI or co
investigator on international research grants (e.g. EC NoE ENFIN).

ordinator of Bioinformatics master course DNA/Protein Structure
Analysis and Prediction.

B4.2 Teusink group

Teusink has joined the IBIVU as chair of Systems Bioinformatics as of 01/09/2008 (0.8fte
FALW and 0.2fte FEW).

He was involved as co
itiator in setting up a Dutch Consortium for Systems Biology of
Lactic Acid Bacteria. He has written the white paper of this Consortium, and was involved as
primary player in all project proposals from this Consortium that were funded to date.
Teusink is c
hairman of the workgroup “Microbial Systems Biology” of the Dutch Society of
Biotechnology (NBV), chairman of the Internal Advisory Board of the NCSB, Program
Leader of the Systems Biology program of the Kluyver Center for Genomics of Industrial
ons (organized through the NGI
funded Netherlands Consortium Systems
Biology). He is member of the management team of the NISB, and coordinator of the


LAB2 consortium in the second SysMO call. Teusink is also advisor for the TI Food
and Nutrition on
modelling and bioinformatics aspects.


Member of the Editorial Board of Molecular Biotechnology

Member of the external evaluation panel for several German BMBF calls, including
the “FORSYS partner” call and the “SYSTEC” call, all related to sy
stems biology

External reviewer for glue projects within the Netherlands Institute for Systems
Biology in Amsterdam

Member of the American Society for Biochemistry and Molecular Biology (ASBMB)


Winner of the Poster Award on the Kluyver Cente
r Symposium 2007 and 2009 (last

Winner of the Poster Award on the NIZO Dairy Conference 2007 (co

Winner of the WCFS Publication price 2006 (first author)

Training and Mobility of Researchers grant EU (Marie Curie) for visiting Prof. Steve
Oliver’s lab at UMIST, Manchester, UK, Oct. 1996

Oct. 1997.


(UD as of 01/11/2008)

Invited lectures at various European universities and research institutes, as well as
international conferences.

PI or co
investigator on (inter)national research
grants (e.g. Kluyver Center, SysMo).

ordinator of MSc programme Systems Biology


Internal evaluation

The Bioinformatics section carries out high quality research in a fairly large number of
directions as detailed above. We feel there is an excellen
t, open and friendly research
atmosphere in the groups, where the PhD students, postdocs and staff communicate closely,
helping each other and collaborating extensively. The same holds for the contacts between the
two groups at IBIVU. Members from both gro
ups share offices, and, given that the Teusink
group is a recent arrival, start collaborating on a number of research project.

The inclusive scientific atmosphere is also appreciated by our students in the Bioinformatics
and Systems Biology master progra
mmes. They find it easy to consult various IBIVU
members active in supervising course assignments, mini
projects and the like. Students
graduating from our master programmes, easily find PhD positions in reputable labs and/or
biotechnology companies, such
as Agendia.

Also the PhD students supervised by us generally manage to complete their PhDs within the
nominal period of four years, and get hired as postdocs in good national or international labs
(e.g. The Broad Institute at Harvard University; Fleming I
nstitute, Athens; CMBI, Nijmegen;
Plant Research International (PRI), Wageningen).


External validation

Heringa has written an overview article discussing the role of Computer Science in
Bioinformatics (in the Dutch language and published in the Aut
omatisering Gids). As chair of
the NBIC Education Committee, he helps setting up education activities for Dutch vocational
education institutes and high schools.


Part of the PhD research project of Bart van Houte has been in collaboration with
on Systems (BDS), an Amsterdam
based SME in biotechnology. We have
developed a number of methods that will help BDS develop purpose
built microarrays and
other biochip technologies.

Thomas Binsl, PhD student in the Heringa group, and his co
promotor Ha
ns van Beek
FALW), together with Koen Verhoeff (VU/VUmc TTO)

have won
the NGI
Venture Challenge 2008
. Their new venture,

the most promising venture
plan, including providing services to the pharmaceutical industry in terms of metabolite
turnover quantification.

Teusink is external advisor of TI Food and Nutrition, a public
private consortium of food
industries and research

institutes, including the universities of Wageningen, Groningen and
Maastricht. Moreover, the Kluyver Center for Genomics of Industrial Fermentations has
strong ties with the (inter)national Biotech industry. Through these links, he has excellent

with industrial partners, evidenced by two STW projects currently running, on top of
projects financed directly through the Kluyver Center.


Researchers and other personnel

As said before, the Bioinformatics section consists of two full professors (
Heringa and
Teusink), which each are embedded in the two founding faculties: Heringa for 0.8fte in FEW
and 0.2fte in FALW, while Teusink acts reciprocally with 0.2fte embedding in FEW and
0.8fte in FALW. There are three assistant professors (Feenstra, Mol
enaar and vacancy),
where Feenstra and vacancy (both 1.0fte FEW) are in the Heringa group, and Molenaar
(1.0fte FALW) in the Teusink group. The vacant UD position is open since about a year: after
Dr. Kleinjung’s departure to London (30/09/2006), the posit
ion has been taken up by Dr.
Marchiori (15/05/2006

28/02/2008). After having lost a candidate who accepted earlier, we
are now actively recruiting again for this position. Kleinjung’s research focus has been on
structural bioinformatics, while Marchiori
has concentrated on bioinformatics data mining
and machine learning. Dr. Feenstra has started as UD in the Heringa group as of 01/01/2007.
Given Dr. Feenstra’s expertise in structural bioinformatics, the new UD currently being hired
should ideally have exp
erience in method and data integration, including data mining and
machine learning.

The table below only lists personnel embedded at the Department of Computer Science
(Bioinformatics section).

Heringa group:

Heringa (0.8fte)

Kleinjung (UD)

Marchiori (


Feenstra (PD, now UD)

Codrea (PD)

Gavai (PD as of 1

Smit (AIO and then PD)

Silva Borges Costa (PD)

Simossis (AIO)

Szklarczyk (AIO)

Binsl (AIO)

Jancura (AIO)


Pirovano (AIO)

Hettling (AIO)

Bonzanni (AIO)


Teusink group:




Oude tabel: moet opnieuw met boevnstaande lijst

We attempt to recruit Ph.D. students from our own international bioinformatics 2
masters program, the new international 1
year bioinformatics
master program, and
bioinformatics programs of other universities. Mobility is stimulated through conference
visits, collaboration in national and international (EC) projects and through visits to other
universities. Postdocs are recruited internationally,

especially from top
universities. For all
personnel, the VU offers various courses, such as English scientific writing and professional
courses for Ph.D. students and for Assistant/Associate Professors. We receive visitors at all
levels of seniority, in
cluding summer
internship master students, Ph.D.students, postdocs and


Resources, funding and facilities


Table 5: Funding at programme level. The full outcome list for this programme is given
separately in Appendix

B.8.1. Heringa group

bioinformatics cluster computer (IBIS1) has been funded partly by the Center for Medical
Systems Biology (CMSB

25%), the Ecogenomics initiative (55%), CvB subsidy to IBIVU
FALW (14%), and CvB subsidy to the IBIVU
FEW (7%).

The IBIVU has been successful
in obtaining research grants in addition to a Central VU
University start
up grant and faculty matching (all 1GS), which includes funding from 2GS
(NGI) but mostly from 3GS (BSIK
NGI, EU). During 2004
2009, the FEW Bioinformatics
section was awarded two 2
GS and seven 3GS grants, representing an equipment grant for a
bioinformatics production cluster computer and research grants for eight Ph.D.
students/postdocs in total. We are actively recruiting to appoint two additional assistant
professors (UD) and on
e postdoc (1GS) and two PhD students at the FEW Bioinformatics

The full list of research funding awarded is as follows:

une 2004:

fellowships Centre for Medical Systems Biology (CMSB): €250K

1 postdoc Systems Biology (Heringa/van Beek


September 2004: Bsik grant Ecogenomics: 1.4 M€ (€700K Bioinformatics


anuary 2005: Bsik fellowship BioRange

175K (4
year PhD studen

ebruary 2005: Bsik fellowship BioRange

175K (4
year PhD student)

May 2005: EC FP6 Network of Excellence “ENFIN” (3
year Postdoc)


September 2005: VU University (Central) Bioinformatics subsidy IBIVU


January 2006: NWO/Horizon Breakthrough
fellowship (1.5
yr Postdoc) €


June 2006:
Bsik fellowship BioRange

175K (4
year PhD student)

October 2006: NGI Centre for Medical Systems Biology fellowship

170K (4
yr PhD

January 2007: NGI Center for Medical Systems Bio
logy fellowship €40K (1
yr PhD)

July 2007: VU/FEW/Computer Science Department collaboration fellowship €85K
(0.5fte 4
year PhD position)

January 2008: VU/FEW/Computer Science Department PhD fellowship €170K (4
year PhD student)

September 2008:
C FP6 Network of Excellence “ENFIN” (2
year Postdoc)


September 2008: NGI Netherlands Bioinformatics Centre (NBIC2) BioRange

220K (3
yr Postdoc, Heringa/van Beek)

September 2008:
EC FP6 Network of Excellence “ENFIN” Experimental Verifi

40K (together with I. Xenarios, Geneva, B. Gottgens, Cambridge)

March 2009: NGI Netherlands Bioinformatics Centre (NBIC2) BioRange

220K (4
yr PhD)

Prospective funding possibilities include

a 2.25

M€ research grant to the IBIV
U Centre within an EU FP7
collaborative project
scale integrating project) VICTOR (Vascular Inflammatory Cells That
Orchestrate Remodelling). This 12

M€ project

submitted to the call FP7





1 Systems biology approaches for
basic biological processes relevant to health and disease. The co
ordinator is

Prof. dr.
Anton J. G. Horrevoets
VUmc. If successful, a consortium comprised of 14 teams
will collaborate i
n this project.


A 460 K€ research grant (one 3
year postdoc for the IBIVU) from NWO
Computational Life Sciences 2009, in collaboration with
Dr. Ir. A.H. de Boer
(FALW, PI) and Dr. C. Jimenez (VUmc, Oncoproteomics) on

Modeling of 14
Isoform Function in the Apoptosis
Network of Tumor Cells”.

B.8.2. Teusink group

Special experimental facilities Teusink group: parallel, flexible fermentor set
up that allows
16 adaptive evolution experiments with individual dynamic control in volumes ranging from
50 mL to 20 L.


Kluyver Centre for Genomics of Industrial Fermentation (NGI) 2PhD, 1 PD as guest,

NBIC (Netherlands Bioinformatics Centre), BioRange project BR11.4, 1 PD, 2009


Computational Life Science grant, 1 PhD 2008

STW project on optimizat
ion of
L. lactis

as cell factory, 1 PhD as guest, 2008

STW project on heterogeneity in
L. lactis

cultures, 1 PD 2009

EC FP7 Network of Excellence UniCellSys, 1 PD, 2009

LAB Systems Biology of MicroOrganisms project, 1 PD location WUR,

Teusink has additionally two PhD positions from 1GS (Van Heerden and vacancy). The latter
position is funded by FEW


Overview of the results

B.9.1. Heringa group key publications

Below are our key publications. We publish the vast ma
jority of our papers in high impact
journals such as
Nucleic Acids Research.


Bonzanni, N., Krepska, E., Feenstra. K.A., Fokkink, W., Kielmann, Th., Bal. H.E.,
and Heringa., J. (2009) Executing multicellular differentiation: quantitative
modelling of
C. elegans

vulval development. Bioinformatics, In press.


Szklarczyk, R., Heringa, J.
, Kosakovsky Pond, S., Nekrutenko, A. (2007). Rapid
asymmetric evolution of a dual
coding tumor suppressor INK4a/ARF locus
contradicts its function
but follows a genomic trend,
Proc. Natl. Acad. Sci. USA


Simossis V.A., Kleinjung, J. and Heringa J. (2005) Homology
extended sequence
alignment. Nucleic Acids Res., 33(3):816


Lin K., Simossis V.A., Taylor W.R. and Heringa J. (2
005) A simple and fast
secondary structure prediction algorithm using hidden neural networks.
Bioinformatics. 21(2):152


Kleinjung J., Romein J., Lin K. and Heringa J. (2004) Contact
based sequence
alignment, Nucleic Acids Research 32(8), 2464

B.9.1. Teusink group key publications


Teusink, B., A. Wiersma, L. Jacobs, R. A. Notebaart, and E. J. Smid. (2009).
Understanding the adaptive growth strategy of L. plantarum on glycerol by in silico
optimisation. PLOS Comp. Biol. in press.



Notebaart, R.
A., B. Teusink, R. J. Siezen, and B. Papp. (2008). Co
regulation of
metabolic genes is better explained by flux coupling than network distance.
Comp. Biol. 4


Teusink, B., and E. J. Smid. (2006). Modelling strategies for the industrial
exploitation o
f lactic acid bacteria. Nat Rev Microbiol 4:46


Teusink, B., A. Wiersma, D. Molenaar, C. Francke, W. M. de Vos, R. J. Siezen, and
E. J. Smid. (2006). Analysis of growth of Lactobacillus plantarum WCFS1 on a
complex medium using a genome
scale metabolic
model. J Biol Chem 281:40041


Teusink, B., F. H. J. Van Enckevort, C. Francke, A. Wiersma, A. Wegkamp, E. J.
Smid, and R. J. Siezen. (2005). In silico reconstruction of the metabolic pathways of
Lactobacillus plantarum: comparing predictions of nutri
ent requirements with those
from growth experiments. Appl Environ Microbiol 71:7253

The table below includes publications produced by staff employed by the FEW
Bioinformatics section.

Oude tabel: moet opn

Table 7: The full outcome list for this programme is given separately (Appendix II)

Below we describe the most important results of our research in sequence analysis, and
functional genomics.

Systems Bioinformatics


grained quantitative mode
lling of C. elegans vulval development using
Petri nets

Together with the Bal and Fokkink groups (both at he Department of Computer
Science) we have developed an infrastructure based upon Petri nets to formally
model multicellular processes in a course
ained way. For the latter aspect, we


have developed the formalism of so
called bounded Petri nets, where the number
of tokens can be constrained, thus enabling the definition of any intermediate
level of concentration or related parameter in between a bool
ean definition
(on/off), as exploited in boolean models, and full chemical quantitation of
compound concentrations used in modeling systems based on sets of ordinary
differential equations (ODEs). The bounded Petri net formalism is now quickly
gaining reco
gnition in the modelling society, owing to our successful modelling
of vulval development in C. elegans. For this biological system, many
in vivo

data is available. We have built a comprehensive multicellular network system,
that turned out to adequately r
eproduce the developmental steps observed in
nature. We were particularly pleased to see that implementation of a newly
discovered micro
RNA gene turned out to be crucially important for adequate
model behaviour (Krepska et al., 2008; Bonzanni et al., 2009

Within the EU project ENFIN (Xenarios group at the Swiss Institute for
Bioinformatics, and the Göttchens group at Cambridge University), we are
currently applying this technology to human haematopoetic stem cells (blood
stem cells), a system known to c
omprise 50 different stem cell forms, where we
are trying to build a network that can adequately reproduce the known gene
activity patterns corresponding to each of the 50 stem cell types.

Sequence Analyis


Multiple Sequence Alignment (MSA):

The Bioin
formatics section is seen internationally as being in the forefront of
multiple sequence alignment. Over the years, we have developed new ways of

based alignment techniques and of integrating protein secondary
structure prediction into multiple

sequence alignment. Our MSA program T
COFFEE (Notredame et al., 2000), which is based on exploiting alignment
transitivity and integration of local and global alignment, has been the most
sensitive and widely used method during a number of years. The last

few years,
the PRALINE webserver has become increasingly popular and receives calls
from 50
100 different users per day. Recently, a new bioinformatics textbook
appeared (Xiong, J.
Essential Bioinformatics.
Cambridge University Press, 2006),
where the mul
tiple alignment suite PRALINE was quoted as “PRALINE is
perhaps the most sophisticated and accurate alignment program available.”
Recently, we developed a solution for the longstanding open problem of multiple
sequence alignment for sequences containing va
rious repeat types (Sammeth and
Heringa, 2006). The resulting method is able to reliably align sequences
containing multiple repeat types, which is an important improvement given that
repeats normally confuse alignment techniques, leading to misalignment.
have recently integrated protein transmembrane segment prediction with multiple
sequence alignment (Pirovano et al., 2008) making use of evolutionary signals in
a soft way, in collaboration with the group of Prof. Michael Gelfand (Moscow
State Universit
Using the TM
aware alignment technology, we are currently
finalising expanding the method by integrating ‘classical’ secondary structure
prediction for soluble proteins in the alignment process (Pirovano and Heringa, in


Genome alignment

We have created a method Aubergene (Szklarczyk and Heringa, 2006) that
exploits transitivity for pairwise alignment of genome sequences while using
information from third genome sequences. The method has extended reliable
alignment to more divergent genome

sequences, and the importance of this has
been well


Protein function
determining sites prediction

We have developed a new entropy
based method to determine function
discriminating sites given a protein multiple sequence alignment that contains


two subgroups of sequences (Pirovano et al., 2006, Feenstra et a., 2007). The
novelty of the algorithm is its modified use of sequence entropy and the fact that
it is not based on amino acid conservation as is the case for most contenders. We
have set up a

small reference database of multiple sequence alignments that
include sequences containing experimentally validated residues. Using this
improved reference set, we demonstrated the superior performance of the new
Sequence Harmony (SH) approach. Based on t
hese results, we moved on by
introducing feature selection technology in the field. This alternative method,
MultiRELIEF, is able to predict functional specificity for multiple (>2) groups,
and exploits a sampling technique that allows its use in large dat


Secondary structure prediction

The group has longstanding experience with developing and assessing secondary
structure prediction protocols. In collaboration with Dr. Kuang Lin (Manchester
University) and Dr. Willie Taylor (NIMR, London), we have d
eveloped a new so
called hidden
Neural Network (HNN) technique that employs a hidden Markov
Model to parse the output from a Neural Network predictor. The resulting
technique, YASPIN (Lin et al., 2005), is particularly accurate at predicting beta
which is the most difficult secondary structure to predict. The technique
is now one of the standard methods in the EVA
server and also our local
YASPIN web server is popular.


Repeats detection

Over the years, the group has gained a reputation in the detec
tion and delineation
of protein sequence repeats. The method REPRO is based on high
scoring non
overlapping local alignments, and employs graph
based clustering for delineation
of the repeat start and stop positions. The REPRO method is particularly suited

for the detection of multiple sets of repeats with varying intervening sequence
fragments. The TRUST method (Sczklarczyk and Heringa, 2004) employs
transitivity and a self
optimising strategy for profile construction, and is
particularly sensitive to deli
neating tandem repeats. The REPRO web server
attracts tens of hits per day, and the TRUST web server draws almost the same
number of hits. The aforementioned technique for multiple alignment of
sequences containing repeats relies on the quality of the REPR

We are also collaborating with the Computer Systems group on optimizing the
performance of our bioinformatics algorithms. The most notable development
here was achieved together with Prof. Bal, where we gained a million
fold speed
ovement for the REPRO algorithm (Romein et al., 2003). We obtained an
magnitude complexity improvement on the sequential algorithm, leading
to a 1000
fold speed
up for average data sizes, and parallelized the improved
algorithm (including MMX vect
or parallelization), yielding another factor 1000.
The improved algorithm is implemented in the REPRO web server.

Functional Genomics


Domain boundary prediction

Parsing a protein sequence into its correct domain structure is one of the most
fundamental ta
sks in functional genomics. We have developed a number of
different algorithms for domain boundary prediction. The SnapDRAGON
protocol (George and Heringa, 2002a) is based on the DRAGON algorithm for
tertiary structure prediction, and derives its accuracy
from a consistency
protocol to delineate domain boundaries over an ensemble (of typically a 100)
different 3D models. Although particularly computationally intensive, this
protocol is recognised as leading to the best boundary predictions, and, for
xample, is now adopted and refined in the Baker group (University of
Washington at Seattle). A salient feature of the SnapDRAGON protocol is that it
is able to delineate discontinuous (or segmented) domains.


A second method, Scooby
Domain, predicts fold
abilility and putative domain
structures by exploiting a partition density function derived from hydrophobic
sequence signals in the CATH protein structure database (George et al., 2005,
Pang et al., 2007).

We have also created a sequence
based method DO
MAINATION that relies on
the PSI
BLAST homology search method (which is currently the most widely
used bioinformatics algorithm). The DOMAINATION method (George and
Heringa, 2002b) parses a query sequence into putative domains based upon a
variety of patte
rns detected in the PSI
BLAST output, and iterates this protocol
until convergence, when no new domain structures can be discerned anymore.
The DOMAINATION method is much faster than the SnapDRAGON technique
and only slightly less accurate. An important re
sult is that the DOMAINATION
protocol improves the performance of homology searching by a sustained 15%
compared to PSI
BLAST. Within the Framework of the ENFIN project (Brandt),
we have recently created a long
awaited webserver (Brandt et al., submitted)
DOMAINATION, while we are developing the method further, for example by
integrating Scooby
Domain into the DOMAINATION protocol..


protein interaction (PPI) prediction

In the context of a NBIC BioRange project (Feenstra) we are exploring new
soscopic modelling techniques to predict protein
protein interactions. The
mesoscopic technique will be intermediate in between full scale atomistic
modeling techniques and sequence
based techniques in that we will project
“cheap” data from our alignment a
nd prediction methods onto spheres
representing protein molecules of which the interaction potential will be
determined. We aim to be able to predict higher
order protein complexes using
this technique. In collaboration with the Huynen group at the CMBI, N
we are also using the evolutionary signal more directly by exploiting multiple
alignment of groups of orthologous and paralogous sequences, where alignment
positions delineated by our SH and MultiRELIEF efforts are likely to represent
positions between putative binding partners.



Within an NBIC BioRange project, a Ph.D. student (Binsl) is modeling metabolic
fluxes based on
isotopomer distributions in metabolic networks
. Prior to
metabolic flux analysis and modeling, metabolit
es have to be identified and
topologies of metabolic pathways have to be reconstructed. Results will become
relevant for groups involved in identification of new metabolites and pathways in
microorganisms and mammalian cells. Moreover, results are expecte
d to benefit
the next stage of pathway and flux analysis in the study of secondary metabolism
in plants.


Genomics research

Rapid asymmetric evolution of a dual
coding tumor suppressor INK4a/ARF locus
contradicts its function but follows a genomic trend (Sz
klarczyk et al., 2007)


Analysis, perspectives and expectations for the research programme

1. Strengths

The Heringa group has a coherent program with a clear focus on sequence analysis, functional
genomics and cellular network reconstruction. It h
as a strong international position and a
number of collaborations within the faculty (Computer Systems, Artificial Intelligence,
genomics, Mathematics), within the Netherlands (NBIC BioRange, NWO
Horizon), and within Europe (ENFIN network). We ha
ve been able to obtain research grants
from NWO, NGI and the EU to fund these collaborative projects. The web interfaces to our


software packages are widely used. Heringa is one of the world leaders in sequence
alignment; for example, the original publica
tion of the T
Coffee multiple sequence alignment
method has been cited 1476 times since publication in 2000 (ISI 16
2009). He was recently
invited to take part in an application to the FP7 Health Call, with prospective funding of 2
million Euro towards t
he IBIVU Centre alone. A Dutch research assistant (S. Smit) at the
University of Boulder selected the Bioinformatics section (Heringa) out of six prospective
institutions for supervising her Ph.D. research, which has mostly been carried out at the
ty of Colorado, Boulder, USA.

The Teusink group has very strong connections within and outside the Netherlands in the field
of lactic acid bacteria. Teusink is a world leader in systems biology. His name is well
in this field, in particular in asso
ciation with the “Teusink”
model of yeast glycolysis, one of
the best
known models in systems biology (cited 174 times since publication in 2000, ISI 16
2009). In recent years, he became one of the European experts in metabolic reconstruction
and constra
based modeling (as exemplified by three invitations in the last FP7 Health
call). Teusink’s group has a strong focus on applying bioinformatics and systems biology
tools to understand metabolic regulation, and combines modeling with experimental work
model construction and validation.

2. Weaknesses

The Heringa group needs to grow to alleviate the teaching load resulting from a full 2
bioinformatics master programme and courses taught at the bachelor level. Critical mass
needs also to be rea
ched to support the Systems Biology initiatives at the VU (MCB) and the
joint Amsterdam initiative (AMOLF, CWI, UvA, VU), particularly given the need for a
comprehensive bioinformatics infrastructure and integration of the various subprojects in this

such as PPI, formal cell modeling and fluxomics. At the moment, we have only two
Ph.D. students actively working in this area, while another postdoc (NBIC BioRange) is being
recruited. Further staffing is also needed to continue the work in domain predict
ion. The two
new assistant professors are expected to ease the current situation once appointed.

The Teusink group currently comprises rather young staff that started more or less at the same
time. The group has to mature into a state where more senior Ph
Ds and young PhDs are
present. There is a need for some postdocs to help in supervising the PhDs. Several projects
have been written specifically to provide for funds for postdocs, two of which have already
been awarded (two postdocs will start around Summ
er 2009), and one is still pending.

3. Opportunities

The Bioinformatics production cluster will be instrumental in raising the level of
computational services to the research community. Industry, notably the new Life Science
Departments of Sun Microsy
stems and Oracle, have expressed interest in a number of our
methods (e.g. PRALINE) and have pledged support via equipment donations, financial
support to workshops, or otherwise. The scientific landscape being shaped by the funded
projects and further ini
tiatives, and the infrastructure provided by FALW groups within the
IBIVU centre, gives rise to many new integrative projects and high
impact developments.
For example, the Systems Biology initiative and TI Pharma collaboration in conjunction with
the oth
er projects in the group could lead to results such as leads for drug design. From the
outset, it was planned to tie in the VUmc as quality of life is a major application area of
Bioinformatics. Although a direct financial injection into the IBIVU centre h
as not been
forthcoming to date, some recent developments (microarray and proteomics groups) in the
new Cancer Center Amsterdam move in the direction of joint research funding as the groups
data mining and machine learning skills are being recognized. Anot
her opportunity in this


regard is a currently evaluated proposal for an Integrated Project (IP) with the European
Commission. This project on vascular and heart disease is co
ordinated by Prof. Horrevoets
from the VUmc. If successful, the IBIVU will receiv
e 2 m€ of research funding for systems
bioinformatics and modelling work on vascular stability and robustness (see also above).

Some established indirect funding from the VUmc involves two PhD students (Binsl and
Hettling) and a postdoc (Gavai), awarded to

Dr. Hans van Beek (VUmc) who are embedded
and housed within the IBIVU centre.

Systems Biology is an important emerging science with many opportunities for scientific
breakthroughs but also for funding. For example, Systems biology is an important compone
in the new Amsterdam Institute for Molecules, Medicine and Systems (AIMMS). Through the
Netherlands Institute for Systems Biology (NISB), based in Amsterdam, many new initiatives
are being started, including a new call for the formation of systems biolo
gy centers (call
expected before Summer 2009).

4. Threats

Hiring good people in Bioinformatics turns out to be difficult, although the international
contacts and the Bioinformatics and Systems Biology master programmes have thus far
enabled us to fill i
n available positions at various levels (Ph.D. students, postdoc, assistant

Apart from finding good candidates, a threat to the Teusink group is overly accelerated
expansion because of the many funding opportunities. Since there is also a lot
of teaching
material that we need to develop in this young field, work loads should not become
prohibitive. Care should be taken on the focus of the group, and the right mix of young and
senior people (i.e. PhDs and postdocs).

5. Analysis

We think that
the group performs well and benefits from: strong leadership and staff (B.1),
strong international reputation (B.4), an important and integrative research mission and
strategy (B.2), a good research focus while maintaining versatility in research direction
(B.3), well
positioned research collaborations (B.6), and adequate and growing external
funding (B.8).

6. Productivity

Given the start
up phase of the group, the relatively small number of research staff, and the
huge investments in setting up the IBI
VU centre and associated education programmes, we
think our productivity thus far has been quite adequate. We emphasize that our policy is to
publish in top journals and not to produce large numbers of overlapping papers. We thus
focus on quality, not qua
ntity. It should also be stressed that Bioinformatics research, unlike
other computer science disciplines, is generally not published in conference proceedings.
Some important bioinformatics conferences (ISMB, RECOMB, ECCB) have agreements with
research journals and publish accepted papers in special issues of these journals.

7. Relevance


The extensive and diverse research collaborations as well as our software distributions and
web interfaces guarantee that the knowledge we produce is adequa
tely disseminated. Our
research has proved useful for the Life Science community.

8. Vitality and feasibility:

As a result of incoming funding the Bioinformatics section is growing rapidly. The large
demand for bioinformatics expertise within life scie
nce departments will lead to an
increasingly pivotal role for the Bioinformatics section, also within the framework of the
convergence between FEW and the FALW. The national and international collaborations and
the embedding of the group in the IBIVU centr
e provide an excellent strategic position to
participate in future large
scale projects, which is important given the continuing tendency of
funding agencies to initiate increasingly large programs.


Appendix 1.

IBIVU organisation structure

To safeguar
d flexible and effective running of the IBIVU and to accommodate its multi
disciplinary research, the following organization structure is implemented (Figure A1).

Governing Board

Governing Board

is formally responsible for the I
BIVU. It delegates the
scientific responsibility to the Directorate of the IBIVU, based on five
year plans. The IBIVU
Directorate reports annually to the Governing Board about progress of the IBIVU. The
Governing Board is responsible for all the IBIVU per

The Governing Board also mediates and eventually decides in cases of disagreement between
Directors and Executive committee when a majority of the latter requests such action. Should
the IBIVU Governing Board be unable to reach consensus within
reasonable time about
issues deemed of major importance, the Governing Board or the Scientific Director resorts to
the University Board (CvB) for a final and binding decision.

The IBIVU Governing Board consists of the Deans of the two founding Facultie
s FEW and
FALW. Additional Deans may become members upon invitation by the Governing Board.



consists of a director and a co
director, who are primarily
responsible. These are the two full professors: Prof. Heringa in I
ntegrative Bioinformatics
(80 % FEW / 20 % FALW) and Prof. Teusink in Systems Bioinformatics (80% FALW / 20%
FEW), respectively. Prof. Westerhoff (FALW) has been co
ad interim
, a position
recently taken over by Prof. Teusink. The Governing Board d
ecides in consultation with the
Executive Committee (see below) regarding changes in the Directorate.

Executive Committee

Executive Committee

is co
responsible for research planning, internal research
evaluation, staffing and financial issues, the ulti
mate responsibility residing with the
Directorate and the Board of Governors (see above). The Executive Committee consists of
the chairpersons of the Thematic Committees (see below) plus the Director and the co

Figure A1.
Organisation structure of the IBIVU.

Governing Board

dean FALW | dean

Board (CvB)

rije Universiteit


Director | co

Executive committee

Thematic Committee chair persons


director of IBIVU. The Executive Committee
appoints these chairpersons on the basis of (i)
scientific excellence, (ii) the balance between the two founding Faculties and further
organizations funding the IBIVU and participating in it, and (iii) the anticipated effectiveness
of the Executive Committ

Thematic Committees

The basic research conducted by the IBIVU is structured in
Thematic Committees
. Each of the
Thematic Committees coordinates, manage and supervise one of the Themes of IBIVU. They
also provide limited support when requested to do

so by the Help Desk (see Section 4.4).
Themes are defined as (i) a collection of related scientific topics for which more than 3
researchers are active (possibly part
time) within the IBIVU (either as IBIVU personnel, or as
external personnel contributed

(often on a part
time basis) to IBIVU, (ii) a collection of
teaching topics and activities effected by IBIVU, and (iii) a collection of other activities, such
as coordinated funding initiatives. The Thematic Committees presented below (see Section
4.1) ar
e structured to allow maximum inclusiveness, synergy and effectiveness of the IBIVU
research, teaching and grant acquisition activities.

Each Thematic Committee consists of all scientists that are active on that theme, either within
the IBIVU or in assoc
iation with the IBIVU. A Thematic Committee proposes a list of
possible chairpersons, out of which the Executive Committee selects the candidate who can
optimally implement the aims of the thematic committees mentioned above, or who has been
most active i
n installing the IBIVU theme. The selected chairperson participates in the
Executive Committee. A Thematic Committee meets at least bimonthly, or more often as

Via its chairperson the Thematic Committee proposes actions to the Executive Committe
which the latter then puts in place, unless there are important reasons not to do so. In the latter
case the Executive Committee justified its decision to the
Thematic Committee
. Actions that
can be proposed by a
Thematic Committee

include the allocati
on of central IBIVU work force
or computational resources to specific research themes, and issuing requests to associated
researchers or institutions regarding activities associated with the research Theme.

Currently the Executive Committee consists of Pr
ofs. Heringa (Director and chairman
IBIVU), Teusink (co
director and secretary IBIVU), Van der Vaart (Math/FEW), Bal
(CS/FEW), Westerhoff (MCF/FALW)Brussaard (ENF/FALW) and Oudega (MC/FALW,

Thematic Committees

are created if important
scientific objectives not covered by
existing Themes are identified, or following acquired research funding in a specific research

Appendix II

Table 7: Overview of the publications of the Programme Bioinformatics



Refereed journ

Simossis V.A. and Heringa J.

(2004) The influence of gapped positions in multiple
sequence alignments on secondary structure prediction methods. Comp. Biol. Chem.
6), 351

Vaccaro L., Cross K.J.,
Kleinjung J.
, Straus S.K. , Thomas D.J., Wharton S.A.,
Skehel J.J., and Fraternali F. (2004) Plasticity of Influenza Haemagglutinin Fusion
Peptides and th
eir Interaction with Lipid Bilayers, Biophys. J. 88, 25

Szklarczyk, R. and Heringa, J.

(2004) Tracking repeats using significance and
transitivity. Bioinformatics

20 Suppl. 1, i311

Simossis V.A. and Heringa J.

(2004) Integrating Protein Secondary Structure
Prediction and Multiple Sequence Alignment, Current Protein and Peptide Science,

Kleinjung J., Romein J.
, Lin K. and
Heringa J.

(2004) Contact
based sequence
alignment, Nucleic Acids Research 32(8), 2464
Full Text (Open Access)

Moradie F., Chan M.P.Y., Telfer M.A., Brandenburg D., Sundermann E.,
Eckey H.,
Kleinjung J.
, Schaefer A. and Jones R.H. (2004) Effect of thyroid
one binding proteins on insulin receptor binding of B1
analogues, Biochemical J. 381(1), 51

E. Marchiori, A. van der Vaart
, G. Meijer, B. Ylstra.
(tool website)
Automatic Breakpoint Identification and Smoothing of Array Comparative Genomic
Hybridization Data. Bioinformatics, 20:3636
3637, 2004.

K. Jong,
E. Marchiori, A. van der Vaart
. Ana
lysis of Proteomic Pattern Data for
Cancer Detection. In Applications of Evolutionary Computing. EvoBIO: Evolutionary
Computation and Bioinformatics.
Springer, pp. 41
51, 2004.

K. Jong,
E. Marchiori
, M. Sebag,
A. van der Vaart
Feature Selection in Prote
Pattern Data with Support Vector Machines. In CIBCB , pp. 41
48, IEEE, 2004.

Kleinjung J., Romein J., Lin K. and Heringa J.

(2004) Contact
based sequence alignment,
Nucleic Acids Research 32(8), 2464

Simossis V.A. and Heringa J.
The influence

of gapped positions in multiple sequence
alignments on secondary structure prediction methods. Comp. Biol. Chem. 28(5
6), 351

Simossis V.A. and Heringa J.
(2004) Integrating Protein Secondary Structure Prediction and
Multiple Sequence Alignment, Cur
rent Protein and Peptide Science, 5(4): 249

Moradie F., Chan M.P.Y., Telfer M.A., Brandenburg D., Sundermann E.,
Eckey H., Kleinjung J., Schaefer A. and Jones R.H.
(2004) Effect of thyroid hormone


binding proteins on insulin receptor binding
of B1
insulin analogues, Biochemical
J. 381(1), 51

Szklarczyk, R. and Heringa, J.

(2004) Tracking repeats using significance and transitivity.
Bioinformatics 20 Suppl. 1, i311

Vaccaro L., Cross K.J., Kleinjung J., Straus S.K., Thomas
D.J., Wharton S.A., Skehel
J.J., and Fraternali F.
(2004) Plasticity of Influenza Haemagglutinin Fusion Peptides and
their Interactions with Lipid Bilayers, Biophys. J. 88, 25

Book chapters:

Heringa J.

(2004) Protein sequence analysis and prediction
of secondary structural features.
In: Compact Handbook of Computational Biology (Konopka A.K. and Crabbe M.J.C., Eds.)
New York, Marcel Dekker, pp. 99

Konopka A.K., Bucher P., Crabbe M.J.C., Crochemore M., Heringa J. Lisacek F.,
Makalowska I., Makalo
wski W., Pesole G., Saccone C., Sagot M.
F., Sarai A., Stadler P.
and Taylor W.R.

(2004) Computational Biology: Annotated Glossary of Terms, In: Compact
Handbook of Computational Biology (Konopka A.K., Crabbe M.J.C., eds.), New York:
Marcel Dekker, pp. 451

Konopka A.K. and Heringa J.

(2004) A dictionary of programs for sequence analysis, In:
Compact Handbook of Computational Biology (Konopka A.K., Crabbe M.J.C., eds.) New
York: Marcel Dekker, pp. 503


Refereed journals:

De Graaf, C., Ver
meulen, N.P.E., and
Feenstra, K.A.

(2005) Cytochrome P450 in Silico: An
Integrative Modeling Approach
J. Med. Chem., 48:2725

Keizers, P.H.J., de Graaf, C., Kanter, Oostenbrink, C.,
, K.A.,

N.P.E., and Commandeur, J.N.M. (2005) Metabolic Regio

And Stereoselectivity Of
Cytochrome P450 2D6 Towards 3, 4
amphetamines: In Silico
Predictions And Experimental Validation.
J. Med. Chem., 48:6117

M. West
Nielsen, E.V. Hogdall,
E. Marchiori
, C.K. Hogdall, C. Schou, N.H.H. Heegaard.
Sample handling for mass spectrometric proteomic investigations of human sera. Analytical
Chemistry, vol. 77, no. 16, 5114
23, 2005.

E. Marchiori
, N. Heegaard, C. Jimenez and M. West
Nielsen. Feature Selection for
Classification with Proteomic Data of Mixed Quality. Proceedings of IEEE Symposium on
Computational Intelligence in Bioinformatics and Computational Biology. CIBCB

, pp. 385
391, 2005.

E. Marchiori
, M. Sebag. Bayesian Learning with Local Support Vector Machines for Cancer
Classification with Gene Expression Data. In Applications of Evolutionary Computing.
EvoBIO: Evolutionary Computation and Bioinformatics.
83, 2005.

Celie, P. H., Kasheverov, I. E., Mordvintsev, D. Y., Hogg, R. C.,
van Nierop, P.
, van Elk, R.,
van Rossum
Fikkert, S. E., Zhmak, M. N., Bertrand, D., Tsetlin, V., et al.
(2005a). Crystal


structure of nicotinic acetylcholine receptor ho
molog AChBP in complex with an alpha
conotoxin PnIA variant.
Nat Struct Mol Biol 12, 582

van Nierop, P.
, Bertrand, S., Munno, D. W., Gouwenberg, Y., van Minnen, J., Spafford, J.
D., Syed, N. I., Bertrand, D., and Smit, A. B. (2005a).
and functional expression
of a family nicotinic acetylcholine receptor subunits in the central nervous system of the
mollusc lymnaea stagnalis. J Biol Chem.

Bosman, L., Lodder, J. C.,
Van Ooyen, A., Brussaard, A. B.
(2005) Role of synaptic
inhibition in
spatiotemporal patterning of cortical activity.
In: Van Pelt, J. Kamermans, M.,
Levelt, C. N., Van Ooyen, A., Ramakers, G. J. A., Roelfsema, P. R. (eds.)
Dynamics and Pathology of Neuronal Networks: From Molecules to Functional Circuits,
ess in Brain Research 147, Elsevier, Amsterdam, pp. 201

Van Ooyen
(2005) Competition in neurite outgrowth and the development of nerve
In: Van Pelt, J. Kamermans, M., Levelt, C. N., Van Ooyen, A., Ramakers, G. J.
A., Roelfsema, P. R. (e
Development, Dynamics and Pathology of Neuronal Networks:
From Molecules to Functional Circuits, Progress in Brain Research 147, Elsevier,
Amsterdam, pp. 81

Kiddie, G., McLean, D.,
Van Ooyen
, A., Graham, B. (2005).
Biologically plausible models
of neurite outgrowth.
In: Van Pelt, J. Kamermans, M., Levelt, C. N.,
Van Ooyen, A.
Ramakers, G. J. A., Roelfsema, P. R. (eds.)
Development, Dynamics and Pathology of
Neuronal Networks: From Molecules to Functional Circuits, Progress in Brain Research 147,

Elsevier, Amsterdam, pp. 67

Roelfsema, P. R.,
Van Ooyen, A.
gated reinforcement learning of internal
representations for classification. Neural Computation 17, 2176

Kleinjung, J.

and Fraternali, F. (2005) POPSCOMP: An autom
ated interaction analysis of
biomolecular complexes, Nucleic Acids Research 33 (Web Server issue), W342

George, R.A., Lin, K., and
Heringa J.

(2005) Scooby
Domain: prediction of globular
domains in protein sequence, Nucleic Acids Res., 33 (Web Serv
er issue), W160

Simossis V.A. and Heringa J.

(2005) PRALINE: a multiple sequence alignment toolbox that
integrates homology
extended and secondary structure information. Nucleic Acids Res., 33
(Web Server Issue), W289

Simossis V.A., Kleinjun
g, J. and Heringa J.

(2005) Homology
extended sequence
alignment. Nucleic Acids Res., 33(3):816

Lin K.,
Simossis V.A.
, Taylor W.R. and
Heringa J.

(2005) A simple and fast secondary
structure prediction algorithm using hidden neural networks. Bioinfo
rmatics. 21(2):152
(Epub 2004 Sep 17).

Heringa J.

(2005) Protein domains. In: Encyclopedia of Genetics, Genomics, Proteomics and
Bioinformatics, Vol. 7 (Subramaniam, Ed.), Section 6: Comparative Methods for Structure
Analysis and Prediction (Taylor, W
.R., Section Ed.); Chapter 68, pp. 3283
3297, Wiley
Interscience, ISBN: 0

Fraternali F. and
Kleinjung J.

(2005) Molecular simulations in structure prediction. In:
Encyclopedia of Genetics, Genomics, Proteomics and Bioinformatics, Vol. 7 (Sub
Ed.), Section 6: Comparative Methods for Structure Analysis and Prediction (Taylor, W.R.,
Section Ed.), Chapter 74, pp. 3346
3352, Wiley Interscience, ISBN: 0


George, R.A., Lin, K. and Heringa J.

(2005) Scooby
Domain: prediction of
domains in protein sequence, Nucleic Acids Res., 33 (Web Server issue), W160

Kleinjung, J. and Fraternali, F.

(2005) POPSCOMP: An automated interaction analysis of
biomolecular complexes, Nucleic Acids Research 33 (Web Server issue), W342

Simossis, V.A. and Heringa, J.

(2005) Local structure prediction of proteins. In:
Computational Methods for Protein Structure Prediction and Modeling. Springer
GmbH, in press.

Simossis V.A. and Heringa, J.

(2005) Secondary structure
guided m
ultiple alignment
quality, Bioinformatics, in press.

Simossis V.A. and Heringa J.

(2005) PRALINE: a multiple sequence alignment toolbox that
integrates homology
extended and secondary structure information. Nucleic Acids Res., 33
(Web Server Issue), W289

Simossis V.A., Kleinjung, J. and Heringa J.
(2005) Homology
extended sequence
alignment. Nucleic Acids Res., 33(3): 816


Pirovano, W.*, Feenstra, K.A.*, and Heringa, J.
(2006). Sequence Comparison by
Sequence Harmony Identifies Subtype S
pecific Functional Sites,
Nucleic Acids Res.,

Feenstra, K.A., Pirovano, W. and Heringa, J.

(2006) Sub
type Specific Sites for SMAD
Receptor Binding Identified
by Sequence Comparison using "Sequence Harmony". In: From
Computational Biophysics to Systems Biology. Forchungszentrum Jülich, Germany; World
Scientific, London,
Vol. 34, pp 73

Graham, B. P.,
Van Ooyen, A.

(2006) Mathematical modelling and numerical simulation of
the morphological development of neurons. BMC Neuroscience 7 (Suppl 1): S9.

Marchiori, E.**, Pirovano, W., Heringa, J. and Feenstra, K.A.**

(2006) A Feature
n Algorithm for Detecting Subtype Specific Sites for Smad Receptor Binding,
ICMLA06 (IEEE), 168
Text preprint

K. Jong,
E. Marchiori, A. van der Vaart
, S. Chin, B. Carvalho, M. Tijssen, P.P. Eijk, P. van
den IJssel, H. Grabsch, P. Quirke, J.J. Oudejans, G.A. Meijer, C. Caldas and B. Ylstra. Cross
atform Array Comparative Genomic Hybridization (array CGH) Meta
Analysis Separates
Hematopoietic and Mesenchymal from Epithelial Tumors. Oncogene , advance online
publication, August 28, 2006; doi:10.1038/sj.onc.1209919.

Patist, J.P., Kowalczyk, W.,
hiori, E.

(2006). Maintaining Gaussian Mixture Models of
Data Streams Under Block Evolution. International Conference on Computational Science,
Springer,pp. 1071
1074, 2006.

Feenstra, K.A.
, Hofstetter, K., Bosch, R., Schmid, A., Commandeur, J.N.M., and V
N.P.E. (2006) Enantioselective Substrate Binding in a Monooxygenase Protein Model by
Molecular Dynamics and Docking
Biophys. J. 91:3206


Lentz, O.,
Feenstra, A.
, Habicher, T., Ha
uer, B., Schmid, R.D., Urlacher, V.B. (2006)
Altering the Regioselectivity of Cytochrome P450 CYP102A3 of Bacillus subtilis by Using a
New Versatile Assay System,
Chem 7:345

Van Nierop, P.
, Bertrand, S., Munno, D. W., Gouwenberg, Y., van Minnen, J., Spafford, J.
D., Syed, N. I., Bertrand, D., and Smit, A. B. (2006). Identification and functional expression
of a family of nicotinic acetylcholine receptor subu
nits in the central nervous system of the
mollusc Lymnaea stagnalis. J Biol Chem 281, 1680

Van Ooyen, A.,

Roelfsema, P. R. (2006). Envisioning the reward. (Preview of Shuler and
Bear, 2006, Science 311: 1606
1609). Neuron 50:188

Steuber, V.,

Willshaw, D.,
Van Ooyen, A.
(2006) Generation of time delays: simplified
models of intracellular signalling in cerebellar Purkinje cells. Network: Computation in
Neural Systems, 17:173

Janulevicius, A., Van Pelt, J.,
Van Ooyen, A.
(2006) Compartme
nt volume influences
microtubule dynamic instability: a model study. Biophysical J. 90:788

Szklarczyk, R., and Heringa, J.
(2006). AuberGene

a sensitive genome alignment tool,
Bioinformatics, 22 (12), 1431

Sammeth, M., and Heringa, J.

6) Global multiple sequence alignment with repeats.
Prot. Struct. Funct. Bioinf. 64, 263

Book Chapters:

E. Marchiori
, C. Jimenez, M. West
Nielsen and N. Heegaard. Robust SVM
based biomarker
selection with noisy mass spectrometric proteomic data. In
Applications of Evolutionary
Computing. EvoBIO: Evolutionary Computation and Machine Learning in Bioinformatics.
Springer, pp. 79
90, LNCS 3907, 2006.

Simossis, V.A., and Heringa, J.
(2005). Local structure prediction of proteins. In:
Computational Metho
ds for Protein Structure Prediction and Modeling (Xu, Y., Xu, D., Liang
J, Eds.), Springer
Verlag, GmbH, in press.


S. Smit
, J. Widmann and R. Knight (2007). Evolutionary rates vary among rRNA structural
Nucleic Acids Res., 35(10): 3339

Szklarczyk, R., Heringa, J.
, Kosakovsky Pond, S., Nekrutenko, A. (2007). Rapid
asymmetric evolution of a dual
coding tumor suppressor INK4a/ARF locus contradicts its
n but follows a genomic trend,
Proc. Natl. Acad. Sci. USA, 104(31): 12807

Chung, W.
Y., Wadhawan, S.,
Szklarczyk, R.
, Nekrutenko, A. (2007). A First Look at
ARFome: Dual
Coding Gene
s in Mammalian Genomes.
PLoS Comp. Biol., 3(5): e91

Pang, C.N.I., Lin, K., Wouters, M.A.,
Heringa, J.
, George, R.A. (2007). Identifying foldable
regions in p
rotein sequence from the hydrophobic signal,
Nucleic Acids Res. 36: 578


Marchiori, E.
, Moore, J.H., Rajapakse, J.C. (Eds.) (2007). Evolutionary Computation,
Machine Learning
and Data Mining in Bioinformatics, 5th European Conference, LNCS
4447, Springer.

Feenstra, K.A., Pirovano, W., Krab, K. and Heringa, J.
(2007). Sequence Harmony:
Detecting Functional Specificity from Alignments,
Nucleic Acids Res., 35: W495

Heringa, J. and Pirovano, W.
Sequence Similarity Searches. In:
Method Express Series
(Dear, P. ed.), Scion Publishing Ltd, Oxfordshire, UK, pp. 39

T.W. Binsl
, K.M. Mullen, I.H.M. van Stokkum
J. Heringa
J.H.G.M. van Beek

FluxSimulator: An R Package to Simulate Isotopomer Distributions in Metabolic Networks;
Journal of Statistical Software, 18(7)

Feenstra, K.A.
, Starikov, E.B., Urla
cher, V.B., Commandeur, J.N.M. and Vermeulen, N.P.E.
(2007). Combining Substrate Dynamics, Binding Statistics and Reactivity to Rationalize
Regioselective Metabolism of Octane and Lauric Acid by CYP102A1 and Mutants,
Sci., 16: 420

M.C. Codrea
, C.R. Jimenez, S. Piersma,
J. Heringa, E. Marchiori

(2007). Robust Peak
Detection and Alignment of nanoLC
FT Mass Spectrometry Data. The Fifth European
Conference on Evolutionary C
omputation, Machine Learning and Datamining. Springer,
LNCS 4447, Pages 35

M.C. Codrea
, C.R. Jimenez,
J. Heringa, E. Marchiori

(2007). Tools for computational
processing of LC
MS datasets: a user's perspective. Computer Methods and Programs in
icine, Elsevier, Volume 86, Issue 3, Pages 281

K.J.A. Vanhoutte, C. Laarakkers,
E. Marchiori
, P. Pickkers, J.F.M. Wetzels, J.L. Willems,
L.P. van den Heuvel, F.G.M. Russel, R. Masereeuw (2007). Biomarker discovery with
TOF MS in human urine as
sociated with early renal injury: evaluation with
computational analytical tools.
Nephrology Dialysis Transplantation, 22(10):2932

Hansen, K., Smit, D. J. A.,
Barkil, A., Van Beijsterveldt, T. E. M.,
Brussaard, A.
, Boomsma, D. I.,
Van Ooyen, A.
, and De Geus, E. J. C. (2007).
Genetic contributions to
range temporal correlations in ongoing oscillations. J. Neuroscience 27: 13882

Krottje, J. K.,
n Ooyen, A.

(2007). A mathematical framework for modelling axon
Bulletin of Mathematical Biology, 69:3


Van Beek, J.H.G.M., Hauschild, A.
C., Hettling, H. and Binsl,


(2008). Robust
Modeling Measurement and Analysis of Human and Animal Metabolic Systems.
Phil. Trans.
, in press.

Pirovano, W., Feenstra, K.A. and Heringa, J.

(2008). The Meaning of Alignment: Lessons
from Structural Diversity.
BMC Bioinformatics, 9(1): 556

Feenstra, K.A., Bastianelli, G. and Heringa, J.

(2008). Predicting Protein Interactions from
Functional Specificity,

From Computational Biophysics to Systems Biology. pp. 89
. Eds. U.H.E. Hansmann, J. Meinke, S. Mohanty and O. Zimmermann, John von Neumann
Institute for Computing, Jülich,
NIC Serie
Vol. 40, 2008


Pirovano, W. and Heringa, J.
(2008). Multiple Sequence Alignment,
Methods Mol. Biol.,
452: 143

Krepska, E.
Bonzanni, N., Feenstra, K.A.
, Fokkink, W., Kielmann, T., Bal, H.E. and
Heringa, J.

(2008). Design Issues for Qualitative Modelling of Biological Cells with Petri
Formal Meth. in Syst. Biol., 5054:48

Brandt, B.W., Heringa, J.

and Leunissen, J.A.M. (2008). SEQATOMS: a web tool for
identifying missing regions in PDB in sequence context,
Nucleic Acids Res., 36: W255
. Ivliev, A.E., 't Hoen, P.A.C., Villerius, M.P., den Dunnen, J.T. and

Brandt, B.W.

(2008). Microarray retriever: a web
based tool for searching and large scale
l of public microarray data,
Nucleic Acids Res., 36: W327

Jancura, P., Heringa, J. and Marchiori., E.

(2008). Dividing Protein Interaction Networks
by Growing Orthol
ogous Articulations. Third IAPR International Conference on Pattern
Recognition in Bioinformatics, PRIB 2008. Springer, 2008.

Jancura, P., Heringa, J. and Marchiori., E.

(2008). Divide, Align and Full
Search for
Discovering Conserved Protein Complexes. P
roceedings of the Sixth European Conference
on Evolutionary Computation, Machine Learning and Datamining in Bioinformatics.
4973, pp. 73

Vajda, I., Van Pelt, J., Wolters, P., Chiappalone, M., Martinoia, S., Van Someren, E. and
Van Ooyen, A.

frequency stimulation induces stable transitions in stereotypical
activity in cortical networks.
Biophysical J., 94: 5028

Pirovano, W., Feenstra, K.A. and Heringa, J.

(2008). PRALINE
: a strategy for improved
multiple alignment of transmembrane proteins,
Bioinformatics, 24(2): 492

Smit, S.
, Rother, K.,
Heringa, J.

and Knight, R. (
2008). From knotted to nested RNA
structures: a variety of computational methods for pseudoknot removal,
RNA, 14(3):410

Lijnzaad, P.,
Feenstra, K. A., Heringa, J.

and Holstege, F
.C.P. (2008) On defining the
dynamics of hydrophobic patches on protein surfaces,
Proteins: Struct. Func. Bioinf.,
114 (2008)

Ye, K.,
Feenstra, K.A, Heringa, J.
, IJzerman, A.P. and


(2008). Multi
RELIEF: a method to recognize specificity determining residues from multiple sequence
alignments using a Machine Learning approach for feature weighting,
24(1): 18

Horner, D.S.,
Pirovano, W.

and Pesole, G. (2008). Correlated substitution analysis and the
prediction of amino acid structural contacts,
Brief. Bioinform
., 9(1): 46

Fujii, T., Ingham, C.J., Nakayama, J., Beerthuyzen, M.M., Kunuki, R.,
Molenaar, D.,

Sturme, M.H.J., Vaughan, E.E., Kleerebezem, M., Vos, W.M. de

(2008) Two homologous
like quorum
sensing systems cooperatively control adherence, cell m
orphology, and cell
viability properties in Lactobacillus plantarum WCFS1. J Bacteriol 190: 7655


Notebaart RA,
Teusink B
, Siezen RJ, Papp B (2008) Co
regulation of metabolic genes is
better explained by flux coupling than network distance. PLOS Comp

Biol 4.

Siezen RJ,

Starrenburg MJC,

Boekhorst J, Renckens B,
Molenaar D
, van Hylckama Vlieg
JET (2008) Genome
scale genotype
phenotype matching of two Lactococcus lactis isolates
from plants identifies mechanisms of adaptation to the plant niche. Appl
Environ Microbiol
74: 424

Stevens, M.J.A., Wiersma, A., Vos, W.M. de, Kuipers, O.P., Smid, E.J.,
Molenaar, D.
Kleerebezem, M.

(2008) Improvement of Lactobacillus plantarum aerobic growth as directed
by comprehensive transcriptome analysis.
Appl Env
iron Microbiol 74: 4776

Book Chapters:

Heringa, J.

(2008). Sequence Similarity. Encyclopedia of Life Sciences

Handbook of
Human Molecular Evolution, John Wiley & Sons Ltd, Chichester, UK.


Brandt, B.W., van Houte, B.,
George, R.A.,

inga, J. (2009).

web server for protein domain delineation using PSI
BLAST. Bioinformatics, submitted.

Smit, S.
, Knight, R. and
Heringa, J.

(2009). RNA structure prediction from evolutionary
patterns of nucleotide composition.
Nucleic Acids


37: 1378

De Groot, P.W.J.,
Brandt, B.W.
, Horiuchi, H., Ram, A.F.J., De Koster, C.G. and Klis, F.M.
(2009). Comprehensive genomic analysis of cell wall genes in
Aspergillus nidulans
Genetics and Biology
, 46(1): S72

Wortman, J.R.
, Gilsenan, J.M.,

Joardar, V., Deegan, J., Clutterbuck, J., Andersen, M.R.,
Archer, D., Bencina, M., Braus, G., Coutinho, P., von Döhren, H., Doonan, J., Driessen,
A.J.M., Durek, P., Espeso, E., Feke
te, E., Flipphi, M., Garcia Estrada, C., Geysens, S.,
Goldman, G., de Groot, P.W.J., Hansen, K., Harris, S.D., Heinekamp, T., Helmstaedt, K.,
Henrissat, B., Hofmann, G., Homan, T., Horio, T., Horiuchi, H., James, S., Jones. M.,
Karaffa, L., Karányi, Z., Ka
to, M., Keller, N., Kelly, D.E., Kiel, J.A.K.W.,Kim, J.
M., van der
Klei, I.J., Klis, F.M., Kovalchuk, A., Krasevec, N., Kubicek, C.P., Liu, B., MacCabe, A.,
Meyer, V., Mirabito, P., Miskei, M., Mos, M., Mullins, J., Nelson, D.R., Nielsen, J., Oakley,
, Osmani, S.A., Pakula, T., Paszewski, A., Paulsen, I., Pilsyk, S., Pócsi, I., Punt, P.J.,
Ram, A.F.J., Ren, Q., Robellet, X., Robson, G., Seiboth, B., van Solingen. P., Specht, T., Sun,
J., Taheri
Taleshi, N., Takeshita, N., Ussery. D., vanKuyk, P.A., Vis
ser, H., van de
Vondervoort, P.J.I., de Vries, R.P., Walton, J., Xiang, X., Xiong, Y., Zeng, A.P.,
, Cornell, M.J., van den Hondel, C.A.M.J.J., Visser, J., Oliver, S.G. and Turner G.
(2009). The 2008 update of the
Aspergillus nidulans

genome annotation: a community effort.
Fungal Genetics and Biology
, 46(1): S2

Brandt, B.W., Heringa, J.

(2009) webPRC: The Profile Comparer for alignment
searching of public domain databases. Nucleic Acids Res., in revision.

Bruggeman, J.,
ringa, J., Brandt, B.W.

(2009) PhyloPars: estimation of missing
parameter values using phylogeny. Nucleic Acid Res., in revision.


Van Houte, B, Heringa J.

(2009) Evolutionary fuzzy clustering of gene expression profiles
in cancer cell lines. Bioinformatic
s, in revision.

Van Houte, B.P.P., Binsl, T.W., Hettling, H., Pirovano, W. and Heringa, J.
(2009) CGH
normaliter: an iterative strategy to enhance normalization of array
CGH data with imbalanced
aberrations. BMC Genomics, in revision.

Bayjanov J,Wels M,
Starrenburg M, van Hylckama Vlieg JET, Siezen RJ,
Molenaar D

(2009) PanCGH: a genotype
calling algorithm for pangenome CGH data. Bioinformatics 25:

Molenaar D
, van Berlo R, de Ridder D,
Teusink B

(2009) The economy of unicellular
growth. Mol Syst

Biol, submitted.

Pastink MI
Teusink B
Hols P
Visser S
de Vos WM
Hugenholtz J
(2009) Metabolic
comparison of lactic acid bacteria; genome
scale model of Streptococcus thermophilus
LMG18311. Appl Environ Microbiol.

Papp B,
Teusink B
, Notebaart RA (
2009) A critical view of metabolic network adaptations.

Teusink B
, Wiersma A, Jacobs L, Notebaart RA, Smid EJ (2009) Understanding the adaptive
growth strategy of
L. plantarum

on glycerol by
in silico

PLOS Comp Biol, in

s M, Bongers RS, Boekhorst J,
Molenaar D
, Sturme M, et al.
(2009) Large Intergenic
like Supermotifs in the Lactobacillus plantarum Genome. J Bacteriol.

Bonzanni, N.
, Krepska, E.,
Feenstra. K.A
., Fokkink, W., Kielmann, Th., Bal. H.E., and
., J.

(2009) Executing multicellular differentiation: quantitative predictive modelling
C. elegans

vulval development. Bioinformatics, in press.

Ph.D. Theses

Simossis, V.A.

From Sequence to Structure and Back Again: an Alignment Tale.
Viva voce

005. Promotor Prof.dr. J. Heringa


Szklarczyk, R.

Information reuse in comparative genomics.
Viva voce

2007. Promotor
Prof.dr. J. Heringa

promotor Dr. Anton Nekrutenko

Penn State University,

Smit, S.

RNA in Formation: Comput
ational Studies on RNA Structure and Evolution.

Promotor Prof.dr. J. Heringa

promotor Dr. R. Knight, University
of Colorado, Boulder, USA.

Notebaart, R.
Integrative Bioinformatics of Metabolic Networks.
Viva voce


Promotores: Prof.dr. R.J. Siezen

RU, Prof.dr. B. Teusink



joint first authors

equal contribution