Methods and software for the analysis and classification of evolutionary and biomedical data

点击翻译按钮获取中文摘要

My research proposal involves four major components: 1. A number of important evolutionary mechanisms can be only explained by the phenomenon of reticulate evolution which allows for additional evolutionary relationships among species compared to the classical model of a phylogenetic tree. First, I propose to develop two algorithms, based on the distance and probabilistic approaches respectively, for the prediction and validation of partial horizontal gene transfer events and the resulting intragenic recombination. The developed algorithms will be applied on a genome scale to estimate the proportion of mosaic genes in genomes of prokaryotes and phages, and used to assess the impact of HGT and intragenic recombination on the resistance of bacteria to antibiotics. 2. Development of two methods and associated software for the analysis and representation of reticulate evolution by hybridization networks. First, I will design a method for computing the smallest number of hybridization events required to explain conflicting gene trees; and, second, I will develop a method for determining whether the relationships among the species at hand should be represented by a phylogenetic tree or by a hybridization network. 3. Investigate the problem of multiple sequence realignment and of joint inference of a phylogeny and a multiple sequence alignment (MSA). The methods developed under this research component will be able to realign long genomic regions and carry out topological rearrangements of the phylogeny in order to optimize the agreement between the final tree topology and the given MSA. 4. Design and implement methods for the analysis and correction of high-throughput screening (HTS) data. I plan to develop two techniques for improving the hit selection procedure in experimental HTS. The proposed methods and related software will allow researchers to minimize the impact of systematic error affecting the selection of potential drug candidates.