Exploiting Homology Information in Nontemplate Based Prediction of Protein Structures

Alfredo Iacoangeli, Paolo Marcatili, Anna Tramontano

Research output: Contribution to journalJournal articleResearchpeer-review


In this paper we describe a novel strategy for exploring the conformational space of proteins and show that this leads to better models for proteins the structure of which is not amenable to template based methods. Our strategy is based on the assumption that the energy global minimum of homologous proteins must correspond to similar conformations, while the precise profiles of their energy landscape, and consequently the positions of the local minima, are likely to be different. In line with this hypothesis, we apply a replica exchange Monte Carlo simulation protocol that, rather than using different parameters for each parallel simulation, uses the sequences of homologous proteins. We show that our results are competitive with respect to alternative methods, including those producing the best model for each of the analyzed targets in the CASP10 (10th Critical Assessment of techniques for protein Structure Prediction) experiment free modeling category.
Original languageEnglish
JournalJournal of Chemical Theory and Computation
Issue number10
Pages (from-to)5045-5051
Number of pages7
Publication statusPublished - 2015
Externally publishedYes

Fingerprint Dive into the research topics of 'Exploiting Homology Information in Nontemplate Based Prediction of Protein Structures'. Together they form a unique fingerprint.

Cite this