Main Page

From BenTal_Group

The Ben-Tal Lab of Computational Structural Biology

Welcome to Nir Ben-Tal's lab at the Department of Biochemistry, The George S Wise Faculty of Life Sciences , Tel Aviv University.

Our research is focused on studying the interplay between protein sequence, structure, motion and function using computational tools. The understanding of these relations provides a molecular dimension to our understanding of protein functions and their involvement in genetic disorders and other diseases. Within the broad fields of structural-bioinformatics and phylogeny, we limit our research to specific niches where structure and motion are difficult to obtain experimentally, and the computations provide data beyond our current knowledge:

Intercellular gap junction channels are formed by the end-to-end docking of two hemichannels, each comprised of a hexamer of connexin subunits. A Cα model (yellow ribbons) for the membrane-spanning domain of the hemichannels was derived by combining the information from a computational analysis of connexin sequences, biochemical studies, and the constraints provided by a 3D map derived by electron cryocrystallography (cryo-electron microscopy, cryo-EM; blue). While individually none of these approaches provided high-resolution information, their sum yielded a model that predicts how connexin mutations (red spheres), which result in genetic diseases such as nonsyndromic deafness and Charcot-Marie-Tooth disease, may interfere with formation of functional channels by disrupting helix-helix packing. (Molecular graphics using AVS software by Michael E. Pique and Mark Yeager.)

Transmembrane proteins

Transmembrane proteins are notoriously difficult to handle experimentally, and their structure determination lags far behind that of water-soluble (globular) proteins. To circumvent these difficulties, we develop and utilize methods to simulate membranes and their interactions with proteins, and to predict the 3-dimensional structure of transmembrane proteins. The figure shows a model structure of the transmembrane domain of the gap junction. The model provides a molecular explanation for almost 30 documented genetic mutations in the protein.

A ConSurf analysis of the evolutionary conservation profile of the platelet glycoprotein alpha-IIb subunit (displayed in CPK models, colored by conservation according to the bar at the bottom), and its interaction with the integrin beta-3 subunit (green ribbon; PDB-1TXV). The interface between the alpha and beta subunits, which is important for proper function of the protein, is evolutionary conserved.

Structural-genomics initiatives

Structural-genomics initiatives aim to produce crystal structures of myriad proteins. This outpouring of structural data is not matched by a similar experimental effort to understand function, and we try to fill this growing gap using phylogenetic analysis. Patches of evolutionarily conserved amino acids that are located in close proximity to each other on the protein surface (see picture on the right) are often important for biological functions; these patches may mediate the association of the protein with other proteins, RNA, DNA or ligand molecules, and may be involved in enzymatic catalysis. We continually develop the Rate4Site algorithm and ConSurf web-server for computation of the evolutionary conservation of the amino acids in proteins.

We develop other computational tools to identify key amino acids that are functionally important, and use them to investigate specific protein families. For example, we develop methodology to look for amino acids that determine specific traits (slight function alterations) within homologous proteins. We also develop methods to search for pairs of amino acids with similar evolutionary history, i.e. pairs of residues that change in tandem and manifest correlated (or compensatory) mutation behavior. We use these methods to investigate selected protein families. As an example, by the use of these and other computational tools, we studied the EGF receptors (also called ErbB and HER), which are involved in breast and lung cancers (EGFR), and suggested a new mechanism of regulation by these proteins. The model provides an explanation, at the molecular level, for the effects of cancer-causing EGFR mutations, and suggests a novel therapeutic venue for EGFR-related cancer.

Systems biology and protein-interaction networks

Ample data on inter-protein interactions are available from high-throughput studies, using, e.g., the yeast two-hybrid method and complex-purification techniques using mass-spectrometry. These relatively recent data are attractive in that they provide vast amounts of interaction data. However, they suffer from a high frequency of false positive interactions and systematic omissions. We aim to bridge the gap between this and structural data as a means to improve the quality of protein networks.