AIS-INMACA: A Novel Integrated MACA Based Clonal Classifier for Protein Coding and Promoter Region Prediction

TL;DR

This paper introduces AIS-INMACA, an innovative classifier combining Cellular Automata and Artificial Immune System techniques, capable of accurately predicting protein coding and promoter regions in DNA sequences, handling large datasets effectively.

Contribution

It presents the first integrated algorithm that simultaneously predicts both promoter and protein coding regions using a novel combination of CA and AIS, achieving high accuracy.

Findings

Average accuracy of 89.6% in predictions

Can handle DNA sequences of various lengths

Effective on large, mixed, and overlapped DNA datasets

Abstract

Most of the problems in bioinformatics are now the challenges in computing. This paper aims at building a classifier based on Multiple Attractor Cellular Automata (MACA) which uses fuzzy logic. It is strengthened with an artificial Immune System Technique (AIS), Clonal algorithm for identifying a protein coding and promoter region in a given DNA sequence. The proposed classifier is named as AIS-INMACA introduces a novel concept to combine CA with artificial immune system to produce a better classifier which can address major problems in bioinformatics. This will be the first integrated algorithm which can predict both promoter and protein coding regions. To obtain good fitness rules the basic concept of Clonal selection algorithm was used. The proposed classifier can handle DNA sequences of lengths 54,108,162,252,354. This classifier gives the exact boundaries of both protein and promoter regions with an average accuracy of 89.6%. This classifier was tested with 97,000 data components which were taken from Fickett & Toung, MPromDb, and other sequences from a renowned medical university. This proposed classifier can handle huge data sets and can find protein and promoter regions even in mixed and overlapped DNA sequences. This work also aims at identifying the logicality between the major problems in bioinformatics and tries to obtaining a common frame work for addressing major problems in bioinformatics like protein structure prediction, RNA structure prediction, predicting the splicing pattern of any primary transcript and analysis of information content in DNA, RNA, protein sequences and structure. This work will attract more researchers towards application of CA as a potential pattern classifier to many important problems in bioinformatics