Page suivante: 1 Introduction
Niveau précédent: Satisfaction de contraintes et
Page précédente: Satisfaction de contraintes et
A basic goal of many experiments in molecular biology is to accumulate, with a minimum of effort, sufficient information concerning an object so that its biologically interesting features can be determined unambiguously. Very often, the data collected from a series of experiments form a fragmentary picture with information of different kinds tainted with errors and uncertainties as to the state of a biological object (a DNA or RNA molecule, a protein...), from which it is desirable to determine features (the sequence of a RNA, the tertiary structure of a molecule...). In this article, several ``classical'' mathematical problems for which methods of artificial intelligence have been used, particularly the framework of constraint satisfaction problems (CSP), are considered and analyzed : physical and genetic mapping ; determination and visualization of molecule structures (RNA, proteins) ; experiment planning (peptide synthesis...). Finally, further developments which could contribute to an improved solving of these problems are discussed.
MOTS-CLéS : Satisfaction de contraintes, biologie moléculaire.
KEY WORDS : Constraint satisfaction, molecular biology.
Page suivante: 1 Introduction
Niveau précédent: Satisfaction de contraintes et
Page précédente: Satisfaction de contraintes et
Copyright(C)1995
INRA
Tous droits réservés