NexusPSO: A novel algorithm to detect transcription factor binding sites

The detection of transcription factor binding sites is a major problem in research in Biology. Methods and computer algorithms can be applied to reduce time complexity and cost of detecting transcription factor binding sites in laboratory experiments. One of the well-known methods commonly used is swarm intelligence. However, errors in detection of transcription factor binding sites can be caused by different binding sites in the same genome sequence. The purpose of this research is to improve the effectiveness and accuracy in the detection of transcription factor binding sites by applying the newly developed pre-processing procedure, Nexus, to Particle Swarm Optimization algorithm (NexusPSO). The accuracy of the NexusPSO algorithm was measured in comparison with other algorithms, using information content (IC) as an indicator, with Escherichia coli data. This study found that NexusPSO is the most accurate method being tested. NexusPSO was then tested using consensus sequences on Saccharomyces cerevisiae and Homo sapiens. NexusPSO showed nearly identical results when compared to DNA footprinting methods.