Identification of Drosophila Promoter Using Positional Differential Matrix and Support Vector Machine from Sequence Data

Abstract

Promoter region plays an important role in controlling gene expression of any living organism. It regulates gene transcription by providing space to the RNA polymerase and transcription factors to bind and interact with. Binding of appropriate transcription initiation complex is determined by the specific promoter sequence carrying gene specific motifs. The promoter recognition process is a part of the complex process where genes interact with each other over time and actually regulates the whole working process of a cell. Thus computational method for identifying promoter is a focal point for researchers. This paper presents an algorithm for identifying Drosophila melanogaster promoter using differential positional frequency matrix between promoter and non-promoter sequences which shows maximum 90.36% tenfold cross validation accuracy. The proposed method exhibits greater accuracy for detecting promoters. Also higher sensitivity and specificity results elucidate that the proposed method is less prone to false negatives and false positives compared to some other existing methods.

Key words:  Drosophila, Promoter, Sequence data

D.O.I. 10.3329/ptcb.v18i2.3394

Plant Tissue Cult. & Biotech. 18(2): 123-130, 2008 (December)