On the integration of protein contact map predictions

Protein structure prediction is a key topic in computational structural proteomics. The hypothesis that protein biological functions are implied by their three-dimensional structure makes the protein tertiary structure prediction a relevant problem to be solved. Predicting the tertiary structure of a protein by using its residue sequence is called the protein folding problem. Recently, novel approaches to the solution of this problem have been found and many of them use contact maps as a guide during the prediction process. Contact map structures are bidimensional objects which represent some of the structural information of a protein. Many approaches and bioinformatics tools for contact map prediction have been presented during the past years, having different performances for different protein families. In this work we present a novel approach based on the integration of contact map predictions in order to improve the quality of the predicted contact map with a consensus-based algorithm.