Tokovenko B. Design and implementation of the in silico methods to identify eukaryotic transcription factor binding sites: ISGF-3 case study

Object: transcription factor binding with specific regulatory sites in gene promoters, change of R. norvegicus genes expression levels in response to interferon alpha (IFN-alpha) treatment. Aim: development of a universal computational method for the identification of transcription factor binding sites (TFBS) and its application for the identification of the genes of primary response to ISGF-3, which is activated after IFN-alpha binding with its specific receptor and represents the dominating interferon-specific signal transduction pathway. Methods: position-weight matrices, hidden Markov models, phylogenetic footprinting, hypergeometric test, exact Fisher test, Bayesian test. Results: based on hidden Markov models, an improved TFBS model was developed. It takes into account not only positional frequencies of the nucleotides, but also higher-order dependencies (namely conserved pairs, triplets etc of neighbour nucleotides). Custom modification of the phylogenetic footprinting method was suggested and implemented. Selecting evolutionary conservative TFBS allows decreasing the number of false-positive results without increasing false-negatives during the TFBS search. For the first time a freely accessible online tool for the identification of TFBS in the eukaryotic gene promoters was developed, using both position-weight matrices and hidden Markov model-based search methods. Ability to conduct genome-wide searches distinguishes the developed tool among similar tools. Using the new tool, a set of putative novel primary interferon-response genes was identified. It was shown that IFN-alpha may regulate expression of a group of genes responsible for the development and functioning of the nervous system.