Structural Bioinformatics - CNRS URA2185  

  HEADDr NILGES Michael /
Dr HUYNH Tru / Dr MALLIAVIN Thérèse / Dr STOVEN Véronique / Dr YERAMIAN Edouard

  Annual Report

The aim of our research is to complement structural studies (X-rays, NMR, Electron microscopy) with in silico studies, to:

better determine and predict three-dimensional structures;

better understand molecular recognition and molecular interactions, with an emphasis on systems with medical relevance.

Our research topics include medically relevant molecular processes (infectious diseases, cancer, and the action of general anesthetics). Collaboration with experimental groups on campus, and our own experimental projects, are of fundamental importance for the group.

We develop strategies for the structural analysis of NMR data to make experimental structure determination more reliable, and allow, for the first time, to obtain an unbiased estimate of quality of an NMR structure. We apply similar methods for structure prediction. We mostly use our homology modeling “pipeline” to answer questions about protein function. Other developments include new probabilistic methods for sequence alignment RNA structure prediction, and gene prediction by physics based genome analysis.

We study the dynamics of protein-protein interactions by docking and molecular dynamics calculations. This provides us new insights into the interplay between protein flexibility and molecular recognition. We could elucidate for the first time the role of conformational entropy in the binding process in detail. The prediction of conformational changes during the binding of two proteins, or a protein and a small ligand, remains an important aim.

The field of protein-ligand interactions has fundamental as well as more applied aspects. In several collaborations with experimental groups we use empirical strategies for ligand docking and virtual screening. Targets include proteins from P. falciparum, T. brucei, T. cruzi, M. Tuberculosis. Our first experiences have been very encouraging. A more ambitious aim is to use free-energy calculations to guide the design of new chemical compounds directly in the binding site of the target.


Figure: Illustration of an important part of the activity of the unit. Starting from structural information for a protein, we use various methods to predict and characterize its interactions with other molecules. The results serve as input for further experimental studies.


Publications 2006 of the unit on Pasteur's references database

Activity Reports 2006 - Institut Pasteur
If you have problems with this Web page, please write to