Multiobjective/Multicriteria Optimization and Decision Support in Drug Discovery

Mohammad Afshar 1 Astrid Lanoue 1 Jean Sallantin 2
2 COCONUT - Agents, Apprentissage, Contraintes
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : We give a brief overview of Multi objective (or Multi criteria) methods in drug design and future directions, focussing on the few published applications in chemoinformatics. Drug design requires simultaneously achieving multiple objectives that are often not independent or can even conflict. However, the multiobjective paradigm has been shown to improve the achievement of multiple objectives when designing combinatorial libraries. It can be used effectively to improve the selection of scalar functions such as QSAR and pharmacophore based models. Future directions include novel rule based methods such as KEM that focus on the identification and analysis of contradictions as a means for exploring optimal (consistent) solution spaces. They show promise when handling large number of objectives in discontinuous search spaces and situations such as multicriteria/multiparametric lead optimization.
Type de document :
Chapitre d'ouvrage
John B. Taylor. Comprehensive Medicinal Chemistry II, 4 (30), pp.767-774, 2007, 978-0-08-045044-5. 〈10.1016/B0-08-045044-X/00275-3〉
Liste complète des métadonnées

https://hal-lirmm.ccsd.cnrs.fr/lirmm-00138626
Contributeur : Jean Sallantin <>
Soumis le : mardi 27 mars 2007 - 09:55:48
Dernière modification le : jeudi 24 mai 2018 - 15:59:23

Identifiants

Collections

Citation

Mohammad Afshar, Astrid Lanoue, Jean Sallantin. Multiobjective/Multicriteria Optimization and Decision Support in Drug Discovery. John B. Taylor. Comprehensive Medicinal Chemistry II, 4 (30), pp.767-774, 2007, 978-0-08-045044-5. 〈10.1016/B0-08-045044-X/00275-3〉. 〈lirmm-00138626〉

Partager

Métriques

Consultations de la notice

300