Skip to Main content Skip to Navigation

Le score local : un outil pour l'analyse de séquences biologiques

Abstract : For any organism, DNA, RNA and proteins information can be considered as long sequences of letters taken from a finite alphabet A. One way to analyze this information is to assign a weight at each letter (an elementary score). Then, we make the sum over each possible segment and search the segment which realizes the maximal score called the local score. Then the problem is to give a level of significance for this local score. We are led to study the distribution of the local score under the null hypothesis : elementary scores are i.i.d. random variables. According to the sign of the mean, the behaviour of the local score is widely different. We determine the asymptotic distribution of the local score when random variables are centered. This work completes the asymptotic study of Karlin & al. and Mercier (the negative expectation). In the area of phase transition we study nurnerically the behaviour of the different approximations. Finally we give the rate of convergence of the cumulative distribution function for the local score over a sequence of length n, as n goes to infinity.
Document type :
File URL :
Complete list of metadata
Contributor : Thèses Ul <>
Submitted on : Thursday, March 29, 2018 - 10:46:10 AM
Last modification on : Tuesday, March 2, 2021 - 5:12:06 PM


  • HAL Id : tel-01746855, version 1



Marie-Pierre Etienne. Le score local : un outil pour l'analyse de séquences biologiques. Mathématiques générales [math.GM]. Université Henri Poincaré - Nancy 1, 2002. Français. ⟨NNT : 2002NAN10289⟩. ⟨tel-01746855⟩



Record views