A Hybrid Anti-notch/Goertzel Model for Gene Prediction in DNA Sequences

Abstract

Accurate detection of the exon regions in DNA sequences using signal processing tools has become a challenge in bioinformatics. In this paper, a hybrid Anti-notch/Goertzel algorithm has been presented for gene selection in DNA sequences. The proposed algorithm has many advantages when compared to other conventional methods. Firstly, it leads to identify the coding protein regions more accurate due to using the Goertzel algorithm which is tuned at the desired frequency. Secondly, faster detection time is achieved. The proposed algorithm is applied on several genes, including genes available in databases BG570 and HMR195 and the results are compared to other methods based on the nucleotide level evaluation criteria. Implementation results showed excellent performance of the proposed algorithm in identifying protein coding regions, specifically in identification of small-scale gene areas.