A Kernel for Protein Secondary Structure Prediction

Yann Guermeur (1) , Alain Lifchitz (2) , Régis Vert (3)

1 MODBIO - Computational models in molecular biology
2 MALIRE - Machine Learning and Information Retrieval
3 LRI - Laboratoire de Recherche en Informatique

Multi-class support vector machines have already proved efficient in protein secondary structure prediction as ensemble methods, to combine the outputs of sets of classifiers based on different principles. In this chapter, their implementation as basic prediction methods, processing the primary structure or the profile of multiple alignments, is investigated. A kernel devoted to the task is introduced, which incorporates high-level pieces of knowledge. Initial experimental results illustrate the potential of this approach.

Mots clés: Bioinformatics; Kernel Methods; Multiclass SVM; Optimal Segmentation; Protein Secondary Structure

Domaines: Apprentissage [cs.LG]; Biochimie [q-bio.BM]; Autre [q-bio.OT]

Yann Guermeur, Alain Lifchitz, Régis Vert. A Kernel for Protein Secondary Structure Prediction. Bernhard Schölkopf, Koji Tsuda, Jean-Philippe Vert. Kernel Methods in Computational Biology, The MIT Press, Cambridge, Massachussetts, pp.193-206, 2004, 0-262-19509-7. ⟨hal-00012701v2⟩