Applying Discrete Fourier Transformation to DNA Motifs Identification

The identification of DNA regulatory motifs (transcription factor binding sites) in co-regulated genes is essential for understanding the regulatory mechanisms. Several approaches are commonly used in searching for motifs, including consensus sequences and position weight matrices. However, identification of regulatory motifs remains a significant challenge. In this paper, a new method called DFTMotifs proposed. Different from traditional methods, DFTMotif applies discrete Fourier transformation (DFT) to motifs discovery. To our knowledge, it is the first motifs-finding method based on DFT. It can get the global optimum more easily than other existing statistical model based methods. We tested its performance on both synthetic and realistic biological data. It proved to be successful in identifying our experimental motifs. Experiments also showed DFTMotif outperformed some popular methods in terms of our experimental data