AISMOTIF-An Artificial Immune System for DNA Motif Discovery

TL;DR

This paper introduces AISMOTIF, an artificial immune system-based algorithm for discovering DNA motifs, which effectively predicts known and novel transcription factor binding sites without prior data knowledge.

Contribution

It presents a novel immune system-inspired algorithm with new weighted scores for motif evaluation, improving motif discovery without needing prior background models.

Findings

AISMOTIF predicts known motifs accurately.

It discovers new motifs effectively.

Performs well compared to eight state-of-the-art algorithms.

Abstract

Discovery of transcription factor binding sites is a much explored and still exploring area of research in functional genomics. Many computational tools have been developed for finding motifs and each of them has their own advantages as well as disadvantages. Most of these algorithms need prior knowledge about the data to construct background models. However there is not a single technique that can be considered as best for finding regulatory motifs. This paper proposes an artificial immune system based algorithm for finding the transcription factor binding sites or motifs and two new weighted scores for motif evaluation. The algorithm is enumerative, but sufficient pruning of the pattern search space has been incorporated using immune system concepts. The performance of AISMOTIF has been evaluated by comparing it with eight state of art composite motif discovery algorithms and found that AISMOTIF predicts known motifs as well as new motifs from the benchmark dataset without any prior knowledge about the data.