We use bioinformatics and theoretical structural biology methods to understand the principles of protein structures and biomolecular interactions. This knowledge is then used to predict protein structures and functions, for drug design and de novo design of new proteins with specific functions. We are particularly interested in:

Proteins with tandem repeats

Genome sequencing projects have revealed the existence of a number of biologically important proteins with tandem arrays of up to 50-residue repeats. [More Info]
These proteins are under-represented in structural databases because their large molecular weight and elongated shapes hamper X-ray crystallography and NMR studies. These difficulties increase the importance of bioinformatics approaches. We pioneered the bioinformatics analysis, classification, structural prediction and modelling of proteins with repeats that fold into a solenoid-like arrangement. Further development of reliable methods for the identification of repetitive protein motifs and ab initio prediction of their 3D structures promises to be a fertile research field in structural bioinformatics. Over the last years many evidences have been accumulated about the high incidence of tandem repeats in the amino acid sequences of virulence factors of pathogens and of some amyloidogenic proteins and prions. Thus, the discovery and structure-function predictions of these domains could lead to the identification of targets for new drugs and vaccines against emerging infectious diseases and to the development of amyloidogenesis inhibitors.

Test our bioinformatics tools for the analysis of tandem repeats in proteins:

  • T-REKS : ab initio identification of tandem repeats
  • PRDB : Protein Repeat DataBase
  • PROFILES : library of sequence profiles

Amyloids and Prions

Several lines of evidence suggest that amyloid fibrils play a causative role in several neurodegenerative diseases, including Alzheimer, Parkinson and Hungtington diseases. [More Info]
Despite a number of efforts, an understanding of the structure and mechanism of amyloid fibril formation remains elusive. This dilemma may be attributed to the fact that methods of high resolution structure determination - X-ray crystallography and NMR spectroscopy - cannot be used on account of the polymeric character and insolubility of the fibrils. The determination of the fibril structure goes throughout suggestions of structural models in the process accumulation of experimental evidences, until a unique model can explain the whole set of data. We are contributing to the progress in this field by developing structural models for the amyloid and prion fibrils. Our ultimate goals are a structure based prediction of amyloidogenic properties of amino acid sequences and rational design of the fibrillogenesis inhibitors.

Design of molecules for biomedical and nanotechnologies applications

Proteins with repeats have many shapes and can form regular oligomers, polymers, 2D and 3D networks. [More Info]
Such versatility opens the way for applications ranging from materials science and nanotechnology to medicine. We use our knowledge on the sequence-structure-function relationships of proteins with repeats to engineer artificial molecules with specific, sought-after features. Examples of our successful designs are “Peptabody”, a high-avidity binding protein, and uniformly soluble alpha-helical nanofibrils.

Protein Structure Motifs

The increasing number of known protein structures gives the opportunity to understand better the relationship between sequence and 3D super-secondary structures (or recurrent 3D folds). [More Info]
We are testing the predictive power of such sequence/structure correlations with the ultimate aim of developing an algorithm for protein structure prediction. Structural knowledge also increases the predictive power of sequence homology search, clustering of protein families and inference of biological function(s) from amino acid sequences. Incorporation of structural information into the sequence database searches is an effective approach to find new evolutionary and functional relationships between proteins. Today, the identification of new protein domains and their relationships is one of the most intense bioinformatics activities. In this area, we are working in close contact with cell biologists and therefore we are mainly focusing on proteins involved in cell cycle regulation.