Using evolutionary information for modelling protein structures and complexes

by Torsten Schwede (University of Basel & SIB Swiss Institute of Bioinformatics)

Wednesday 30 Mar 2016, 14:00 → 15:00 Europe/Rome

Aula Conversi (Dip. di Fisica - Edificio G. Marconi)

By exploiting information from evolutionary related proteins, protein structure homology modelling techniques can provide 3-dimensional models for proteins when direct experimental structures are not available. Easy to use web-based modelling services allow comparative modelling techniques today to be used routinely in a broad spectrum of applications in life science research. Two decades ago, SWISS-MODEL pioneered the field as the first fully automated protein modelling service on the Internet, and remains today on! e of the most widely used modelling servers. Ultimately the accuracy of a model determines its suitability for specific applications. Model quality estimation (QE) is therefore an essential component of structure modelling to provide realistic model confidence estimates. In this presentation, I will discuss the SWISS-MODEL expert system (http://swissmodel.expasy.org/) for automated protein structure modelling, as well as our efforts in model quality assessment and estimation.