Top-Down Hierarchical Ensembles of Classifiers for Predicting G-Protein-Coupled-Receptor Functions

摘要: Despite the recent advances in Molecular Biology, function of a large amount proteins is still unknown. An approach that can be used prediction protein consists searching against secondary databases, also known as signature databases. Different strategies applied to use signatures proteins. A sophisticated inducing classification model for this prediction. This paper applies five hierarchical methods based on standard Top-Down and one method new named Ensembles - combination classifiers three different functional datasets employ signatures. The algorithm presented slightly better results than other algorithms, indicating combinations improve performance models.