E-mail Print PDF

The SIB Molecular Modelling Group


Group Leaders: Olivier Michielin - Vincent Zoete

Olivier Michielin - Vincent Zoete


The Molecular Modelling group (MMG) studies mechanisms of molecular recognition, in particular protein-protein or protein-small ligand interactions. The group develops and employs molecular modelling techniques such as homology, molecular dynamics, docking and free energy simulations. Most efforts are concentrated on the development of new small molecule inhibitors of important targets for cancer therapy, as well as the design of optimized peptides vaccines or T Cell Receptor (TCR) sequences for cancer immunotherapy.


Rational design of small molecule inhibitors

In silico structure-based ligand design is becoming a very attractive alternative to high throughput in vitro methods.  We have developed the EADock docking program, which uses a very accurate and universal scoring function to provide a good description of molecular interactions from small fragments up to complete ligands. An efficient conformational search engine has been designed based on an evolutionary algorithm. We are currently using a fragment-based approach to design specific inhibitors of important cancer targets.

Rational optimization of peptides and TCR sequences for immunotherapy

Specific cellular immune responses are based on the recognition by cytotoxic T lymphocytes of immunogenic peptides presented in the context of the class I Major Histocompatibility Complex (MHC). Peptide modifications that increase the affinity for the MHC are of considerable importance for peptide-based vaccinations. Alternatively, TCR sequence modifications designed to improve recognition of a given p-MHC complex represent a very attractive approach since these modified sequences can be incorporated in the patient’s lymphocytes using viral vectors and used in an adoptive transfer setting. There is therefore a need for free energy calculation methods not only to dissect TCR-p-MHC interactions, but also to interpret the effect of a mutation and guide peptide and/or TCR modifications.

Projects and Services

EADock / SwissDock

The EADock program has been designed to provide a flexible solution for general purpose docking, lead optimization and fragment-based drug design. It takes into account the solvation free energy by means of the Generalized Born model, as well as the flexibility of both the ligand and the protein. EADock has been benchmarked on several test sets and obtained a very robust docking success rate of around 80%.

Using EADock, several inhibitors of important cancer targets like integrins, PPARs or indoleamine-2,3-deoxygenase have recently been designed. Low micromolar, and in some cases high nanomolar, inhibitors were obtained.

In parallel, SwissDock, a Web interface, is being developed to allow researchers around the world to perform EADdock based docking simulations, as well as fragment-based drug design and lead optimization.

Hybrid Quantum Mechanical/Molecular Mechanical (QM/MM) Docking

We developed a hybrid quantum mechanical/molecular mechanical (QM/MM) on-the-fly docking algorithm, which addresses the challenges of treating polarization and metal interactions in docking. The algorithm is based on our classical docking algorithm Attracting Cavities and relies on the CHARMM force field. We tested the performance of this approach on three very diverse data sets: (1) the Astex Diverse set of common noncovalent drug/target complexes formed both by hydrophobic and electrostatic interactions; (2) a zinc metalloprotein data set, where polarization is strong and ligand/protein interactions are dominated by electrostatic interactions; and (3) a heme protein data set, where ligand/protein interactions are dominated by covalent ligand/iron binding. Redocking performance of the on-the-fly QM/MM do cking algorithm was compared to the performance of classical Attracting Cavities and other popular docking codes. The results demonstrated that the QM/MM code preserved the high accuracy of most classical scores on the Astex Diverse set, while it yielded significant improvements on both sets of metalloproteins at moderate computational cost.

Computer-Aided Drug Design for Immuno-Oncology

Immuno-oncology has delivered exciting results over the last decade, characterized by substantial and long-term clinical benefit for cancer patients. A promising strategy to further improve clinical outco mes consist in inhibition of the hemoprotein indoleamine 2,3-dioxygenase 1 (IDO1), which plays a major role in tumor-induced immunosuppression. We are rationally designing small molecule modulators of thi s target using different computational and experimental techniques. Our fragment-based computational drug-design effort led to the discovery of highly efficient IDO1 inhibitors, the most active being of n anomolar potency both in enzymatic and cellular assays, while showing no cellular toxicity and a high selectivity for IDO1 over related enzymes.

Structural analysis of TCR immune repertoire

The group is developing a system for producing high quality 3D models of TCR-p-MHC complexes for any given experimental sequence. The approach uses homology modelling and ab initio predictions for the complementary determining region loops of the TCR. Comparisons between the TCR-p-MHC models allow conserved structural motifs that are not apparent at the sequence level to be detected. This approach is also made publicly available through a Web interface.

Rational optimization of peptides and TCR sequences for cancer immunotherapy

Rational optimization of peptides and TCR sequences requires the determination of the binding free energy component associated with each side chain involved in the interaction between the TCR and the p-MHC complex. We have developed or adapted several free energy calculation methods to address this question, including an MMGB-SA approach and alchemical free energy calculations using thermodynamical integration. These methods are now used to optimize several TCR sequences that have been shown to play an important role in a patient’s immune responses. They will be used for adoptive transfer therapies. Similarly, tumour-specific peptides optimized with these techniques will be used for vaccination trials.

Websites for Further Information


google analytics