Herramientas bioinformáticas para el estudio del ADN: caracterización de genes, estudio filogenético y detección de secuencias funcionales

Resumen

In this study, bioinformatics tools for DNA sequence analysis were developed. The main objective was to explore existing tools and propose new uses for them.

The starting point of this work was the Nandy representation, which transforms a nucleotide chain into a set of points in the plane, according to the order of the bases that compose it. This representation was devised in order to highlight the local relative abundance of nucleotides and to differentiate between regions with large variations.

This work has explored two new applications: the geometric and multifractal characterization of the representations and their application to the study of phylogeny. The parameters obtained from the characterization have been related to different characteristics of the sequences, for example, the presence of introns or the kingdom to which they belong. Also, methods have been proposed to estimate phylogenetic relationships between individuals through geometric comparison of DNA sequences of homologous genes. This was attempted in other studies, but new approaches have been taken.

For the localization of non-coding sequences with functionality in long DNA strands, a probabilistic approach is proposed. The starting hypothesis is that, in vertebrates, the distribution of nucleotides in zones which are not under selection pressure, have constant relative base frequencies.

For the development of this project, a database of 10,400 DNA sequences from different organisms, used both in sequence characterization and phylogenetic study, was prepared.

The results obtained show the potential of these tools, both in the field of phylogeny and in the localization of functional sequences, which can complement the existing ones.