Predicting enzyme class from protein structural parameters and bagging predictors.

Short Abstract: In this work we present a new method to classify enzymes that uses the STING_DB physical-chemical parameters and Bagging predictors. By building models based on "decision tree" and "neural network", we obtained an accuracy of 74% on average. These results outperform the similar models proposed in literature.

Saved in:
Bibliographic Details
Main Authors: YAMAGISHI, M. E. B., OLIVEIRA, S. R. M., BORRO, L. C., SANTOS, E. H., JARDINE, J. G., VIEIRA, F. D., MAZONI, I., NARCISO, M. G., KUSER-FALCÃO, P. R., NESHICH, G.
Other Authors: MICHEL EDUARDO BELEZA YAMAGISHI, CNPTIA; STANLEY ROBSON DE MEDEIROS OLIVEIRA, CNPTIA; LUIZ C. BORRO; EDGARD HENRIQUE DOS SANTOS, CNPTIA; JOSÉ GILBERTO JARDINE, CNPTIA; FÁBIO DANILO VIEIRA, CNPTIA; IVAN MAZONI, CNPTIA; MARCELO GONCALVES NARCISO, CNPTIA; PAULA REGINA KUSER FALCÃO, CNPTIA; GORAN NESHICH, CNPTIA.
Format: Anais e Proceedings de eventos biblioteca
Language:Ingles
English
Published: 2006-08-17
Subjects:Parâmetros estruturais da proteína, Parâmetros de Sting_DB, Bioinformática, Proteina, Enzima, Proteins, Enzymes, Bioinformatics,
Online Access:http://www.alice.cnptia.embrapa.br/alice/handle/doc/9310
Tags: Add Tag
No Tags, Be the first to tag this record!