Research
IN-AIS-MACA: Integrated Artificial Immune System based Multiple Attractor Cellular Automata For Human Protein Coding and Promoter Prediction of 252bp Length DNA Sequence
Gene prediction involves protein coding and promoter predictions. There is a need of integrated algorithms which can predict both these regions at a faster rate. Till date, we have individual algorithms for addressing these problems. We have developed a novel classifier IN-AIS-MACA, which can predict both these regions in genomic DNA sequences of length 252bp with 93.5% accuracy and total prediction time of 1031ms. This classifier will certainly create intuition to develop more classifiers like this.
A Fast Multiple Attractor Cellular Automata with Modified Clonal Classifier Promoter Region Prediction in Eukaryotes
DNA is a very important component in a cell, which is located in the nucleus. DNA contains lot of information. For DNA sequence to transcript and form RNA which copies the required information, we need a promoter. So promoter plays a vital role in DNA transcription. It is defined as “the sequence in the region of the upstream of the transcriptional start site (TSS)â€. If we identify the promoter region we can extract information regarding gene expression patterns, cell specificity and development. So we propose a novel fast multiple attractor cellular automata (MACA) with modified Clonal classifier for promoter prediction in eukaryotes. We have used three important features like TATA box, GC box and CAAT box for developing this classifier. The proposed classifier is tested with datasets from Eukaryotic Promoter Database, EPDnew which is a collection of promoters of human, mouse, zebrafish and D.melanogaster. In training phase of the classifier 100% specificity was obtained. In testing phase 84.5% sensitivity and 92.7% specificity was achieved in an average. The time taken to predict the promoter region of length 252 in an average is .7 nano seconds.
