Bonvin Lab

NWO compute grant for 10M CPU hours to support the HADDOCK service

2026-02-12T00:00:00+00:00

The WeNMR services, provided by our group at Utrecht University since 2010, have been at the forefront of delivering essential computational resources to the global life sciences and structural biology communities, supporting the research activities of more than 77,000 registered users to date. These services, backed by multiple European projects and initiatives, have made substantial contributions to scientific advancement. Among the flagship tools offered is the HADDOCK platform, widely used for the modelling of biomolecular interactions.

The provision of these computational services entails considerable resource demands, amounting to approximately 5,500 CPU years annually—far exceeding the capacity of the Utrecht team’s local infrastructure. Through access to High Throughput Computing (HTC) resources within the European Open Science Cloud (EOSC) under the EGI federation, in which the Dutch national grid participates, the HADDOCK service has been able to effectively support a large and continuously expanding international user base.

Through the renewal of an NWO-EW computing grant (grant number 2025.054), we have secured continued access to Dutch HTC grid resources, in the form of 10 million compute hours provided by SURFSara and Nikhef, ensuring the sustained operation of our computational services.

NWO compute grant for 10M CPU hours to support the HADDOCK service was originally published by Alexandre Bonvin at Bonvin Lab on February 12, 2026.

New EU project funded on sustainable computing - ENSURE

2026-02-10T00:00:00+00:00

The Bonvin group is a partner in the recently funded HORIZON-INFRA-2025-01-TECH-01 ENSURE project led by EGI, which focuses on reducing the environmental impact of large-scale digital Research Infrastructures (RIs). RIs consume significant energy and resources as they generate and process vast amounts of data. The project will develop standardized metrics, monitoring tools, digital twins, and AI-based models to track and predict environmental impacts, while integrating sustainability into workload management and reporting frameworks.

Our HADDOCK webservice is a use case, with the aim of reducing the CO2 footprint of HADDOCK calculations by a smart distribution of jobs across the EGI/EOSC high throughput compute resourses.

New EU project funded on sustainable computing - ENSURE was originally published by Alexandre Bonvin at Bonvin Lab on February 10, 2026.

EMBO poster prize for Anna Engel

2025-12-11T00:00:00+00:00

Congratulations to Anna Engel for winning the poster prize at the EMBO workshop on Computational structural biology which took place in Heidelberg Dec. 2-5. Anna presented her poster entitled: “Protein-Quest: Finding structural treasures in the protein jungle” – work supported by NWO (Dutch Research Council) and the Netherlands eScience Center. Want to know more about what this is about? Check https://github.com/haddocking/protein-quest.

EMBO poster prize for Anna Engel was originally published by Alexandre Bonvin at Bonvin Lab on December 11, 2025.

HADDOCK3 antibody-antigen Colab tutorial

2025-09-23T00:00:00+00:00

We have released a Jyputer notebook running on Colab resources that demonstrates the use of HADDOCK3 for predicting the structure of an antibody-antigen complex using information about the hypervariable loops of the antibody and a loose definition of the epitope determined through NMR experiments.

To access and launch the notebook visit https://github.com/haddocking/haddock3/tree/main/notebooks

Or launch it directly on Colab here!

Note that that a command line version of this tutorial is also available from our education pages at: https://www.bonvinlab.org/education/HADDOCK3/

And a similar tutorial using instead our HADDOCK2.4 webserver is available from: https://www.bonvinlab.org/education/HADDOCK24/HADDOCK24-antibody-antigen-basic/

HADDOCK3 antibody-antigen Colab tutorial was originally published by Alexandre Bonvin at Bonvin Lab on September 23, 2025.

GANANA - EU-INDIA Partnership for Scientific High-Performance Computing

2025-09-23T00:00:00+00:00

Our group is a partner in GANANA, a new EU project funded under HORIZON-EUROHPC-JU-2023-INCO-06. GANANA aims at establishing a long-term partnership collaboration between European HPC centers of excellence (including the BioExcel Center for Computational Biomolecular Research to which the Bonvin group belongs – bioexcel.eu) and Indian institutions. GANANA’s objectives are to strengthen the links between EU and Indian research communities and develop selected leading software packages by extending their functionalities, optimising their HPC performance and deploying those on Indian computing resources, improving usability and data integration.

Within GANANA the UU group is leading the life sciences activities that are centered around two European software: Gromacs for molecular dynamics (KTH Stockholm) and the HADDOCK integrative modelling software developed in Utrecht. Check the interview with Alexandre Bonvin:

GANANA - EU-INDIA Partnership for Scientific High-Performance Computing was originally published by Alexandre Bonvin at Bonvin Lab on September 23, 2025.

Inside WeNMR’s Journey with EGI

2025-07-31T00:00:00+00:00

Now serving over 66,000 users in 174 countries, WeNMR’s success is built on user-friendly tools like the HADDOCK server, strong support and training, and a commitment to scientific excellence.

Check out the interview with Alexandre Bonvin about empowering structural biology published in the EGI Magazine.

Inside WeNMR’s Journey with EGI was originally published by Alexandre Bonvin at Bonvin Lab on July 31, 2025.

The long path to therapeutic antibodies

2025-06-13T00:00:00+00:00

Merus N.V., a biotech company based in Utrecht, the Netherlands, has recently announced the FDA approval of BIZENGRI®, a bispecific antibody (zenocutuzumab-zbco) as the first and only treatment indicated for adults with pancreatic adenocarcinoma or non–small cell lung cancer (NSCLC) that are advanced unresectable or metastatic and harbor a neuregulin 1 (NRG1) gene fusion who have disease progression on or after prior systemic therapy.

The approved therapeutic incorporates Merus’ patented DEKK technology that promotes the efficient production of antibodies comprising CH3 domains having particular modifications, including bispecific antibodies that contain two different heavy chains.

3D view of a bifunctional antibody.
The two different heavy chains are indicated in wheat and light blue colors, while the similar light chains are shown in lime yellow.

DEKK refers to four charged substitutions (D:Asp, E:Glu, K:Lys) that were introduced into the Fc domain of the antibody. The general approach followed to develop these modifications and the crystal structure of the DEKK antibody Fc domain (PDB entry 5NSC) were published in 2017 in the Journal of Biological Chemistry authored by Merus and Utrecht University researchers (De Nardis et al. 2017)*. The article describes, among others, how the BioExcel HADDOCK software was used, more than 10 years ago, to evaluate a variety of substitutions. HADDOCK was among the tools used by the scientist to develop and assess mutations, using among others HADDOCK to score the various combinations. This led to identification of a number of candidates that potentially favored the asymmetric assemblies of the heavy chains over the symmetric ones. These combinations of variations were then extensively evaluated and characterized by a battery of biochemical and biophysical methods leading ultimately to the the DEKK modifications, which are integrated into BIZENGRI®.

C. De Nardis,LJ.A. Hendriks,E. Poirier, T. Arvinte, P. Gros, A.B.H. Bakker, J. de Kruif. A new approach for generating bispecific antibodies based on a common light chain format and the stable architecture of human immunoglobulin G1. J. Biol. Chem. 292, 14706-14717 (2017).

This development work, incorporating HADDOCK as part of integrated in silico and benchwork experimentation nicely illustrates how molecular modelling software like the ones at the core of the EuroHPC BioExcel Center of Excellence for Computational Biomolecular Research can contribute to the early stages in the long path leading to the development of therapeutic antibodies.

The long path to therapeutic antibodies was originally published by Alexandre Bonvin at Bonvin Lab on June 13, 2025.

Two postdoc positions in computational structural biology

2025-02-12T00:00:00+00:00

Introduction

We are inviting applications for two postdoctoral researcher positions with interest and expertise in computational structural biology and research software development. These positions are part of GANANA, a collaborative European/Indian project aimed at establishing a long-term partnership collaboration between European HPC centers of excellence (including the BioExcel Center of Excellence for Computational Biomolecular Research to which the Bonvin group belongs) and Indian institutions.

Within GANANA our UU group is leading the life sciences activities that are focused on two European software: Gromacs for molecular dynamics (KTH Stockholm) and HADDOCK, our integrative modelling software developed in Utrecht.

Your job

In this international collaborative project, you contribute to the further development of the HADDOCK integrative modelling platform, taking ownership in one or more of the following GANANA areas:

Development of a framework for Integrative Modelling of Biomolecular Complexes.
Extension of HADDOCK’s capabilities for the modelling of modified nucleic acids.
Development of Improved AI models for protein-protein interactions.
Creation of a framework for AI-driven Immunogenic Peptide Prediction.
Deployment and optimisation of HADDOCK on Indian HPC Cloud resources.

You will actively engage with both the European and Indian research teams, with opportunities for collaborative visits to the partner labs. This role offers unique opportunities to develop not only your research and software skills, but also your collaboration, management and leadership skills as Utrecht leads the life sciences efforts within GANANA.

Your qualities

This position is perfect for someone who enjoys solving complex challenges, likes working in an international and multidisciplinary environment, and wants to make an impact in computational structural biology. Ideally, you meet several or all of the following criteria:

a PhD in chemistry, physics, biology, computational sciences or related fields;
a track record in computational structural biology;
demonstrable programming and software development experience (please provide a link to your online portfolio or code samples);
experience with Linux and high-throughput and high performance computing;
strong communication skills in English, both oral and written;
the ability to operate in an international setting.

Our offer

a chance to be part of a team of excellent researchers in the field of computational structural biology;
an exciting role within a large European/Indian collaboration;
A position for 2 years;
A gross monthly salary ranging from €4.060 to €4.383 in scale 10;
8% holiday bonus and 8.3% end-of-year bonus;
A pension scheme, partially paid parental leave, and flexible terms of employment based on the Collective Labour Agreement Dutch Universities.

In addition to the terms of employment laid down in the CAO NU, Utrecht University has a number schemes and facilities of its own for employees. These include agreements on facilitating professional development, leave and sports and cultural activities, as well as discounts on software and other IT products. We also offer access to additional employee benefits through our Terms of Employment Options Model. In this way, we encourage our employees to continue investing in their growth. For more information, please visit Working at Utrecht University.

About us

A better future for everyone. This ambition motivates our scientists in executing their leading research and inspiring teaching. At Utrecht University, the various disciplines collaborate intensively towards major strategic themes. Our focus is on Dynamics of Youth, Institutions for Open Societies, Life Sciences and Pathways to Sustainability. Sharing science, shaping tomorrow.

Working at the Faculty of Science means bringing together inspiring people across disciplines and with a variety of perspectives and backgrounds. The Faculty has six departments: Biology, Pharmaceutical Sciences, Information & Computing Sciences, Physics, Chemistry and Mathematics. Together, we work on excellent research and inspiring education. We do so, driven by curiosity and supported by outstanding infrastructure. Visit us on LinkedIn and discover how you can become part of our community.

Your position will be embedded in the Computational Structural Biology group, part of the NMR section at Utrecht University that aims at gaining atomic-level insight into complex (bio)molecular systems in vitro, in situ and in silico. The group belongs to the Bijvoet Center for Biomolecular Research and the Chemistry Department.

The team’s research focuses on the development of reliable bioinformatics and computational approaches to predict, model and dissect biomolecular interactions at atomic level. It has a long history of software and web services developments, with HADDOCK as flagship software and web portal that serves a large international community of users (>55000 users from >155 different countries).

More information

For more information, please contact Prof. Alexandre Bonvin at a.m.j.j.bonvin@uu.nl

Do you have a question about the application procedure? Please send an email to science.recruitment@uu.nl

Apply now

As Utrecht University, we want to be a home for everyone. We value staff with diverse backgrounds, perspectives and identities, including cultural, religious or ethnic background, gender, sexual orientation, disability or age. We strive to create a safe and inclusive environment in which everyone can flourish and contribute.

If you are enthusiastic about this position, just “apply here”! Please enclose:

your letter of motivation;
your Curriculum vitae, including clear demonstration of your programming and computing skills
the names, telephone numbers, and email addresses of at least two references;

If this specific opportunity isn’t for you, but you know someone else who may be interested, please forward this vacancy to them.

Some connections are fundamental – Be one of them

Two postdoc positions in computational structural biology was originally published by Alexandre Bonvin at Bonvin Lab on February 12, 2025.

2025 EMBO integrative modelling course

2025-01-16T00:00:00+00:00

Registration is open for the 2025 edition of our EMBO practical course on Integrative Modelling of Biomolecular Complexes. It will take place Sept. 28 - Oct. 3, 2025 in Izmir, Turkey.

Experimental structural studies of interactions can be costly, low-throughput, and challenging. Therefore, computational methods aiming to model protein complexes are particularly valuable. The accurate modeling of multi-scale protein interactions often requires the use of experimental data during modeling. Researchers will thus benefit tremendously from learning how to use integrative modeling methods and AI tools, which will open the possibility for hypothesis generation and testing.

Within this context, this EMBO Practical Course is designed to teach computational approaches and recent AI developments for predicting how proteins interact with other biomolecules or ligands. We provide theoretical and applied background on state-of-the-art algorithms for modeling biomolecular complexes, the use of low- and high-resolution experimental data, molecular dynamics information, coevolution-based interface predictions, as well as AI-based structure prediction techniques. By uniting different computational expertise under the umbrella of this course, we create a platform to stimulate discussions on modeling challenging systems, such as molecular machines.

Roughly half of our course will consist of practical sessions where the participants will run computations on interesting biological problems. To encourage interaction between the tutors and participants and stimulate discussions, the participants will be prompted to present their own research, both in flash presentations and poster sessions and to bring their own research problems to dedicated troubleshooting sessions.

Check the website for the full programme and for registration information. The registration deadline is June 15th.

2025 EMBO integrative modelling course was originally published by Alexandre Bonvin at Bonvin Lab on January 16, 2025.

HADDOCK workshop at ISGC2023

2023-02-21T00:00:00+00:00

We are organising a HADDOCK integrative modelling workshop at Academia Sinica in Taipei as a satellite event of the International Symposium on Grid and Cloud computing. The workshop will take place on March 20th. Registration information can be found at the ISGC2023 website.

This workshop (detailed programme here) will consist of lectures and computer tutorials during which some of the recent HADDOCK developments will be discussed and demonstrated in hands-on computer tutorials. In particular we will introduce HADDOCK3, the new modular version of HADDOCK which represents a redesign of the HADDOCK2.X series, implementing new ways to interact with the HADDOCK sub-routines and offering more customization to the end user.

For the computer practical, participants are expected to bring their own laptop and register in advance to the event we created on NMRBOX as we will be making use of those cloud resources for the HADDOCK3 workshop. Participants are encouraged to bring their own problems to get advice on the best modelling strategy.

HADDOCK workshop at ISGC2023 was originally published by Alexandre Bonvin at Bonvin Lab on February 21, 2023.

HPC innovation award for deep learning work

2020-12-15T00:00:00+00:00

In the context of a SURF Open Innovation Lab collaboration for deep learning enhanced HPC applications, the consortium of several universities (WUR, Radboud, Utrecht and Leiden universities) and SURFSara, led by Caspar van Leeuwen and Axel Berg, have received for their work an 2020 HPC Innovation and ROI Award.

Part of the work involves our work on using deep learning to classify crystallographic interfaces as biological or crystal artifact. The approach builds on our DeekRank project in collaboration with the Netherlands e-Science Center.

Read the details in our whitepaper:

C. van Leeuwen, D. Podareanu, V. Codreanu, M. Cai, A. Berg, S. Portegies Zwart, R. Stoffer, M. Veerman, C. van Heerwaarden, S. Otten, S. Caron, C. Geng, F. Ambrosetti, A.M.J.J. Bonvin. Deep-learning enhancement of large scale numerical simulations. Whitepaper (2020).

HPC innovation award for deep learning work was originally published by Alexandre Bonvin at Bonvin Lab on December 15, 2020.

Targeting membrane complexes

2020-12-10T00:00:00+00:00

Membrane proteins are among the most challenging systems to study with experimental techniques in the structural biology field. The increasing number of deposited structures of membrane proteins has opened the route to modeling their complexes by methods such as computational docking. We have developed a new protocol for the modeling of membrane-associated protein assemblies, which successfully integrates a simulated membrane calculated by the MemProtMD database into a first docking step by LightDock with a flexible final refinement with HADDOCK. This protocol should shed light on the still dark fraction of the interactome consisting of membrane proteins and will be extremely useful for drug discovery, antibody design and personalized medicine.

Nature Comm. 11, 6210 (2020)
https://doi.org/10.1038/s41467-020-20076-5

Targeting membrane complexes was originally published by Alexandre Bonvin at Bonvin Lab on December 10, 2020.

Live demo of the HADDOCK EOSC portal

2020-05-28T00:00:00+00:00

At the occasion of the EOSC-Hub week conference that took place online (May 18-20, 2020), Alexandre Bonvin gave during closing plenary session an interview and live demo of the HADDOCK thematic service. The demo shows how EOSC enables the WeNMR services to support thousands of researcher worldwide in the field of structural biology. WeNMR is linked to the Instruct-ERIC ESFRI. The structural biology community has focused its efforts in the last months on providing answers to fight the Sarc-Cov-2 pandemic. The demo shows how the WeNMR services contribute to this research.

Live demo of the HADDOCK EOSC portal was originally published by Alexandre Bonvin at Bonvin Lab on May 28, 2020.

HADDDOCK COVID drug repurpusing

2020-04-17T00:00:00+00:00

The novel coronavirus (SARS-CoV-2) that has emerged from Wuhan, China in December 2019 has spread to almost all countries in the World causing a dramatic number of deaths. The current absence of antiviral treatment against the SARS-CoV-2 urges the scientific community to accelerate the drug discovery research process.

One way to identify potential treatments and to be able to administer it swiftly is to focus on drug repurposing studies, i.e. to investigate the SARS-CoV-2 antiviral potential of drugs that have already been approved for human use.

Proteins that are crucial for the survival and replication of the virus are the most attractive targets for such studies. We have focused on the SARS-CoV-2 main protease (3CLpro) that plays an essential role in the virus replication process by screening ~2000 approved drugs against this particular protein. We used for that our HADDOCK software, powered by the HTC resources of EOSC/EGI.

More details and initial results are now available online here.

HADDDOCK COVID drug repurpusing was originally published by Alexandre Bonvin at Bonvin Lab on April 17, 2020.

High throughput computing NWO grant to support computational structural biology through the Utrecht WeNMR services

2020-03-18T00:00:00+00:00

HADDOCK (High Ambiguity Driven protein-protein DOCKing) is an integrative platform developed at Utrecht University for the modelling of biomolecular complexes. HADDOCK distinguishes itself from other docking methods in the fact that it can use a large variety of experimental and bioinformatics data to drive the docking process. HADDOCK can deal with a large class of modelling problems including protein-protein, protein-nucleic acids, protein-peptide and protein-ligand complexes.

Since 2008 HADDOCK has been offered as a web portal, freely accessible to non-profit users (wenmr.science.uu.nl) as part of the WeNMR services. HADDOCK is also a thematic service in the European Open Science Cloud marketplace. This strategy has maximized the impact and valorisation of the research of our group (and the impact of NWO funding). Besides the application software, the service also provides automated pre- and post-processing, compute, temporary storage and job scheduling and monitoring for running the application, so that researchers do not need to worry about application porting and compute infrastructure. The portal is heavily used with >15500 registered users from >110 countries (see statistics), and receives a high number of citations.

We have obtained a NWO-ENW computing grant that allows us to maintain our services provisioning to serve the scientific community at large and maximise the impact of Dutch-funded research. Access to High Throughput Computing (HTC) grid resources is key. Not only does it allow to serve an international community, but it is also central to the research of the Utrecht group since significant computational recourses are required to keep developing new methods and model increasingly more complex and large macromolecular assemblies. The relevance of modelling biomolecular complexes is further highlighted by the fact that several major pharma companies are using our software. Our online services have also find their way into teaching as we see recurring registrations from students of various international universities.

SURFSara and the Dutch Dutch National Grid Initiative have always supported the computing required by the Utrecht portals, providing resources for more than 70% of the jobs that run last year. This computing grant allows us to ensure the continued operation of our portals by renewing our access to the Dutch HTC grid resources. For this, we have been granted 8 million SBU wall clock hours (4M per year).

High throughput computing NWO grant to support computational structural biology through the Utrecht WeNMR services was originally published by Alexandre Bonvin at Bonvin Lab on March 18, 2020.

Panos Koukos successfully defends his PhD

2020-03-09T00:00:00+00:00

On February 26th, 2019, Panos Koukos successfully defended his PhD Thesis (read more below).

Another doctor in the group!

Well done and congratulations

Panos thesis entitled Integrative modelling of biomolecular complexes: From small to large focusses on biomolecular docking, a field of computational structural biology which concerns itself with solving 3D puzzles describing how biomolecules interact to form complexes. More specifically, it zooms into complexes involving proteins embedded in the membrane surrounding every cell, and complexes between small compounds and protein receptors of pharmaceutical interest.

The thesis begins by describing the current state of the art in integrative modelling approaches, with a particular emphasis on docking. Integrative modelling refers to computational methods that make use of diverse sources of data to guide the simulations. Chapters two and three are dealing with membrane protein complexes, first establishing a benchmark dataset and then describing a method to model those by docking. Chapters four to six are devoted to small molecule docking with the former two describing our successful participation in an international blind docking challenge, and the latter a novel method making use of shape information to guide the docking of small molecules into their receptors. The thesis ends with critical perspectives and areas of interest for future developments.

Panos Koukos successfully defends his PhD was originally published by Alexandre Bonvin at Bonvin Lab on March 09, 2020.

Review on Integrative Modelling

2019-11-27T00:00:00+00:00

Check out our fresh of the press review on Integrative Modelling of Biomolecular Complexes.

It also discusses the modelling of membrane complexes.

Congratulations to Panos Koukos for a great job!

Review on Integrative Modelling was originally published by Alexandre Bonvin at Bonvin Lab on November 27, 2019.

2020 EMBO integrative modelling course

2019-11-27T00:00:00+00:00

Registration is open for the third edition of our EMBO practical course on Integrative Modelling of Biomolecular Complexes. It will take place Mayt 10-15, 2020 in Izmir, Turkey.

This course will present docking-based computational methods and related bioinformatics approaches, aiming at predicting how proteins interact with other biomolecules or ligands. Lectures will provide the theoretical background on state-of-the-art algorithms for sampling and scoring docking models, describe the use of low- and high-resolution information, and conservation- and coevolution-based interface prediction methods. Roughly half of the course wil consist of practical sessions where the students will run computations on interesting biological problems.

Check the website for the full programme and for registration information.

A poster can be downloaded here

2020 EMBO integrative modelling course was originally published by Alexandre Bonvin at Bonvin Lab on November 27, 2019.

Aye Aye Captain

2019-05-01T00:00:00+00:00

We have been since a long time contributing various software, including our HADDOCK flagship software, to SBGrid. Read the recent story about a developer’s tale highlighting our research and HADDOCK.

https://sbgrid.org/software/tale/aye-aye-captain

Aye Aye Captain was originally published by Alexandre Bonvin at Bonvin Lab on May 01, 2019.

Coconuts for drug design

2019-04-30T00:00:00+00:00

Our recent paper in the Journal of Chemical Information and Modelling introduces a computational approach based on metadynamics simulations called EDES – Ensemble-Docking with Enhanced-sampling of pocket Shape – that allows to generate holo-like conformations of proteins only exploiting their apo structures. This is achieved by defining a set of collective variables that effectively sample different shapes of the binding site, ultimately mimicking the steric effect due to ligands.

Our article features on the cover of JCIM with the analogy of a coconut showing a well open binding pocket compared to a filled watermelon not leaving any space for the ligand.

We assessed the method on three challenging proteins undergoing different extents of conformational changes upon ligand binding. In all cases our protocol generates a significant fraction of structures featuring a low RMSD from the experimental holo geometry. Moreover, ensemble docking calculations using those conformations yielded in all cases native-like poses among the top ranked ones.

Read the full story at:

A. Basciu, G. Malloci, F. Pietrucci, A.M.J.J. Bonvin and A.V. Vargiu. Holo-like and druggable protein conformations from enhanced-sampling of binding pocket shape. J. Chem. Inf. and Mod. 59, 1515–1528 (2019).

Coconuts for drug design was originally published by Alexandre Bonvin at Bonvin Lab on April 30, 2019.