Abstract : We propose here to learn automata for the characterization of proteins families to overcome the limitations of the position-specific characterizations classically used in Pattern Discovery. We introduce a new heuristic approach learning non-deterministic automata based on selection and ordering of significantly similar fragments to be merged and on physico-chemical properties identification. Quality of the characterization of the major intrinsic protein (MIP) family is assessed by leave-one-out cross-validation for a large range of models specificity.
https://hal.inria.fr/inria-00070340
Contributeur : Rapport de Recherche Inria <>
Soumis le : vendredi 19 mai 2006 - 20:10:33
Dernière modification le : vendredi 13 janvier 2017 - 14:20:27
Document(s) archivé(s) le : dimanche 4 avril 2010 - 20:59:32
François Coste, Goulven Kerbellec. A Similar Fragments Merging Approach to Learn Automata on Proteins. [Research Report] RR-5672, INRIA. 2005, pp.17. <inria-00070340>
