Model Independent Method for Acceptor Splice Sites Prediction in DNA Sequences

Signal Processing plays a very important role in the annotation of genome data. It helps to find out different structural features present in the DNA sequences like exonic regions, intronic regions, untranslated regions, promoter regions, CpG islands, etc. The detection of the exonic regions (protein coding regions) is very important for accurate gene prediction. The accurate identification of exonic regions is associated with the prediction of acceptor and donor splice sites. In this work, adaptive short time Fourier transform (ASTFT), period-3 measure, and principal component analysis (PCA) based model independent method for the acceptor site prediction has been proposed. The performance of the proposed method has been compared with the windowed discrete Fourier transform (WDFT).