Title :
Prediction of protein coarse contact maps using recursive neural networks
Author :
Frasconi, Paolo ; Vullo, Alessandro
Author_Institution :
Dipt. di Sistemi e Informatica, Univ. di Firenze, Italy
Abstract :
Tools for predicting topological features such as residue contacts can serve as a crucial step towards prediction of protein folding from sequence. In this paper we focus on the prediction of contacts between secondary structure segments. We introduce a new machine learning approach for scoring candidate contact maps and we show that the predicted score can be effectively used to guide a graph search algorithm.
Keywords :
biology computing; molecular biophysics; molecular configurations; neural nets; proteins; candidate contact maps scoring; graph search algorithm guidance; machine learning approach; protein coarse contact maps prediction; protein folding prediction; protein sequence; recursive neural networks; secondary structure segments; Bioinformatics; Data structures; Feedforward neural networks; Genomics; Learning systems; Machine learning; Machine learning algorithms; Neural networks; Predictive models; Proteins;
Conference_Titel :
Molecular, Cellular and Tissue Engineering, 2002. Proceedings of the IEEE-EMBS Special Topic Conference on
Print_ISBN :
0-7803-7557-2
DOI :
10.1109/MCTE.2002.1175038
