Detection of Regulatory Motif in Eukaryotes by Self Organizing Map Neural Networks

Abstract

Transcriptional control is governed by the actions of a large number of proteins, called transcription factor. Gene is regulated by the binding of transcription factor on the regulatory motifs known as Transcriptional factor binding sites. Genes are regulated by activity of short DNA sequences (regulatory motif) of size 6-12 that resides in close proximity to the co-regulated genes in the genome. Many diseases are caused by defects in gene regulation, rendering the identification of regulatory sequences is an important task. Many of the Current motif finding systems uses clustering based algorithms. This assumption has some limitations because sequence signals have distinct properties and varies in count. This project aims at identifying old and new regulatory motifs in DNA Sequences using Self-Organizing Map (SOM) and Neural Networks. This system is based on a novel intra-node soft competitive procedure to achieve maximum discrimination of motifs from background signals in datasets. The intra-node competition is based on an adaptive weighting technique on two different signal models to better represent these two classes of signals. System is developed as a Motif analysis tool using that researchers will find motif sequences for several real and artificial dataset.

Keywords: A - Adenine, G - Guanine, C - Cytosine, T - Thymine, Regulatory Motif, Self Organizing Neural Network, Position Specific Scoring Matrix, Hybrid Analysis Model