Skip to Main content Skip to Navigation
Theses

Le transcriptome : un domaine d'application pour les
statistiques, de nouveaux horizons pour la biologie

Abstract : The analysis of transcriptomic data requires statistical methods to provide reliable conclusions.
Amongst the huge amount of available methods, biologists may have difficulties to choose the most
appropriate one for their needs. The existing criterions to compare different methods are either
incomplete or use criteria that are not biologically relevant.
The organisation of bacterial genomes offers a biologically relevant criterion to compare the methods
independently of the goal of the experiment: the operons. We have developed a protocol based on this
criterion and compared some classical methods: PCA, ICA, t-test and ANOVA.
Furthermore, meta-analysis of transcriptome data is currently developed. These meta-analyses allow
the study of new biological fields such as the chromosomal organisation of gene expression. We have
analysed three bacteria, B. subtilis, E. coli and S. meliloti and have revealed long-range correlations of
expression in all organisms, whatever the gene studied.
Document type :
Theses
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-00067855
Contributor : Anne-Sophie Carpentier <>
Submitted on : Tuesday, May 9, 2006 - 2:54:32 PM
Last modification on : Saturday, June 6, 2020 - 4:35:16 PM
Long-term archiving on: : Saturday, April 3, 2010 - 11:20:20 PM

Identifiers

  • HAL Id : tel-00067855, version 1
  • PRODINRA : 252517

Collections

Citation

Anne-Sophie Carpentier. Le transcriptome : un domaine d'application pour les
statistiques, de nouveaux horizons pour la biologie. Sciences du Vivant [q-bio]. Université d'Evry-Val d'Essonne, 2006. Français. ⟨tel-00067855⟩

Share

Metrics

Record views

627

Files downloads

4647