An efficient sliding window strategy for accurate location of eukaryotic protein coding regions.

The sliding window is one of important factors that seriously affect the accuracy of coding region prediction and location for the methods based on power spectrum technique. It is very difficult to select the appropriate sliding step and the window length for different organisms. In this study, a novel sliding window strategy is proposed on the basis of power spectrum analysis for the accurate location of eukaryotic protein coding regions. The proposed sliding window strategy is very simple and the sliding step of window is changeable. Our tests show that the average location error for the novel method is 12 bases. Compared with the previous location error of 54 bases using the fixed sliding step, the novel sliding window strategy increased the location accuracy greatly. Further, the consumed CPU time to run the novel strategy is much shorter than the strategy of the fixed length sliding step. So, the computational complexity for the novel method is decreased greatly.