3530 articles – 5253 references  [version française]

hal-00616640, version 1

Estimating the Class Posterior Probabilities in Protein Secondary Structure Prediction

Yann Guermeur () 1, Fabienne Thomarat () 1

PRIB 2011 (2011) ???

  • 1:  ABC (Apprentissage et Biologie Computationnelle) (LORIA)

  • CNRS : UMR7503 – INRIA – Université Henri Poincaré - Nancy I – Université Nancy II – Institut National Polytechnique de Lorraine (INPL) France

Bibliographic reference

  • Type of document: Peer-reviewed conferences/proceedings
  • Subject:
    Computer Science/Bioinformatics
    Life Sciences/Quantitative Methods
  • Title: Estimating the Class Posterior Probabilities in Protein Secondary Structure Prediction
  • Abstract: Support vector machines, let them be bi-class or multi-class, have proved efficient for protein secondary structure prediction. They can be used either as sequence-to-structure classifier, structure-to-structure classifier, or both. Compared to the classifier most commonly found in the main prediction methods, the multi-layer perceptron, they exhibit one single drawback: their outputs are not class posterior probability estimates. This paper addresses the problem of post-processing the outputs of multi-class support vector machines used as sequence-to-structure classifiers with a structure-to-structure classifier estimating the class posterior probabilities. The aim of this comparative study is to obtain improved performance with respect to both criteria: prediction accuracy and quality of the estimates.
  • Fulltext language: English
  • Book title: Actes de PRIB 2011
  • Audience: international
  • Publication date: 2011-11
  • Page, identifiant, ...: ???
  • Conference or book title: PRIB 2011
  • Conference date: 2011-11
  • Country: Netherlands
  • Keyword(s): protein secondary structure prediction – multi-class support vector machines – class membership probabilities
 
  • hal-00616640, version 1
  • http://hal.archives-ouvertes.fr/hal-00616640
  • oai:hal.archives-ouvertes.fr:hal-00616640
  • From: Yann Guermeur <>
  • Submitted on: Tuesday, 23 August 2011 15:38:15
  • Updated on: Tuesday, 23 August 2011 15:38:15