Skip to Main content Skip to Navigation
Theses

Estimations précises de grandes déviations et applications à la statistique des séquences biologiques

Abstract : To establish lists of words with unexpected frequencies in random sequences, for instance in a molecular biology context, one needs to quantify the exceptionality of families of word frequencies. We study large deviation probabilities of multidimensional word counts in Markov models and hidden Markov models. To prove these results, we establish Edgeworth-like expansions on multidimentional fonctionals of finite Markov chains. We use those theorems to get lists of words with unexpected frequencies in the genomic sequences of Escherichia Coli and Bacillus Subtilis.
Document type :
Theses
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-00008517
Contributor : Pierre Pudlo <>
Submitted on : Wednesday, February 16, 2005 - 5:19:35 PM
Last modification on : Wednesday, November 20, 2019 - 2:38:19 AM
Long-term archiving on: : Friday, April 2, 2010 - 9:12:29 PM

Identifiers

  • HAL Id : tel-00008517, version 1

Collections

Citation

Pierre Pudlo. Estimations précises de grandes déviations et applications à la statistique des séquences biologiques. Sciences du Vivant [q-bio]. Université Claude Bernard - Lyon I, 2004. Français. ⟨tel-00008517⟩

Share

Metrics

Record views

327

Files downloads

1001