Utilization of Machine-Learning Methodologies in Order to Understand Complex Evolutionary and Functional Links among Bacterial Genomes - Archive ouverte HAL Access content directly
Conference Papers Year : 2013

Utilization of Machine-Learning Methodologies in Order to Understand Complex Evolutionary and Functional Links among Bacterial Genomes

Olivier Poirion
  • Function : Author
  • PersonId : 943199
Bénédicte Lafay
Connectez-vous pour contacter l'auteur

Abstract

We are searching for evolutionary trends among genome maintenance-related genes present on the replicon sets (i.e., chromosomes and plasmids) of bacterial genomes. Traditional bioinformatic and phylogenetic methods are not adapted to large scale and high-dimensional study. We thus developed a semi-supervised analytical pipeline relying on data-mining methodologies. Generic unsupervised (SOM, K-means, Bayesian networks) and supervised (SVM,decision trees, boosting) classification methods were combined with specific bioinformatic algorithms based on sequence homology search (BLAST). Through this approach, important evolutionary processes could be characterized among genome-integrated plasmids and chromosomes. We here report on the inherent difficulties (input data bias, high-dimensional analysis, noise) and the applied methodology, and conclude on the significance of the data-mining methodology in knowledge discovery.
Fichier principal
Vignette du fichier
IFCS_2013-Poirion_Lafay.pdf (27.29 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-00999074 , version 1 (03-06-2014)

Identifiers

  • HAL Id : hal-00999074 , version 1

Cite

Olivier Poirion, Bénédicte Lafay. Utilization of Machine-Learning Methodologies in Order to Understand Complex Evolutionary and Functional Links among Bacterial Genomes. 2013 conference of the International Federation of Classification Societies (IFCS) : 'United through Ordination and Classifi- cation', Jul 2013, Tilburg, Netherlands. p. 203. ⟨hal-00999074⟩
467 View
47 Download

Share

Gmail Mastodon Facebook X LinkedIn More