Implementation and Development of the DSP Algorithm for Splice Site Prediction in DNA Sequence

Abstract

Now a days genomics signal processing play an important role in bioinformatics area and one of the important problem of this area is gene identification. Various techniques are developed for gene prediction over the past many years. In order to apply DSP tools to DNA sequence, symbolic nucleotides of DNA must be transformed into a numerical sequence and these are affecting the performance of the algorithms. This project report gives the full study of numerical mapping for protein coding region using short time Fourier transform further the study is focused on exons (protein coding region) as to find its correct location in DNA sequence. Further the evaluation parameter that is (Sensitivity, Specificity and the area under the curve) are considered and are calculated. Twenty mappings are implemented and are compared with each other by comparing their area under the curve. This work provides an accessible study and a review of various Mapping techniques and prediction of exons with Digital signal processing tools and further to find exact location of protein coding region. Further this work provides the splice site of the sequence and calculation of score for Acceptor region with PSSM method.