Analysis of DNA-dimer distribution in retroviral genomes using a Bayesian networks induction technique based on genetic algorithms

Authors:
Ramiro Garza-Domínguez;Antonio Quiroz-Gutiérrez
Affiliations:
Centro de Tecnologias de Informacion, Universidad Autonoma del Carmen, Ciudad del Carmen, Campeche, Mexico;Facultad de Quimica, Universidad Autonoma del Carmen, Ciudad del Carmen, Campeche, Mexico
Venue:
MICAI'07 Proceedings of the artificial intelligence 6th Mexican international conference on Advances in artificial intelligence
Year:
2007

Citing 4
Cited 0

Structure Learning of Bayesian Networks by Genetic Algorithms: A Performance Analysis of Control Parameters

IEEE Transactions on Pattern Analysis and Machine Intelligence
Artificial intelligence: a new synthesis

Artificial intelligence: a new synthesis
Practical genetic algorithms

Practical genetic algorithms
Fuzzy Logic

Fuzzy Logic

Quantified Score

Hi-index	0.00

Visualization

Abstract

Since DNA-dimer analysis has demonstrated to provide a very conserved pattern that has been suggested as a genome signature, in this paper we present a computational study of DNA-dimer distribution in a collection of Retroviral genomes. This analysis is based on two main steps: the generation of the target dataset, in this step, the DNA-dimer distribution variables are calculated and transformed to categorical data using Fuzzy Sets. And the induction of a Bayesian Network from the dataset. This induction technique is based on Genetic Algorithms. We have found interesting causal relationships between the DNA-dimer distribution variables and a set of chemical variables. These results could provide new directions in future Retroviral genomic investigations. The computational methodology presented in this paper has demonstrated to be an interesting tool for the study and the analysis of genomic sequences.