AliBaba2: context specific identification of transcription factor binding sites.

摘要: Currently, prediction of transcription factor binding sites is widely done using matrices collected from literature. This leads to several problems. We cannot actively control the conservation matrices, we systematically use all available, do not know which were used and discarded in matrix construction, compare evaluate easily, detect redundancy sensitivity specificity. So are lacking during identification process. In this paper a method overcome these problems proposed. It assumed that each site has an unknown context determines its sequence. idea constructing specific for sequence analysing. To so have regard as general process, starting at dataset known ending with potential new site. such process presented. Besides overcoming mentioned problems, implementation also reaches significantly higher accuracy than current approaches. Evaluations analysing TRANSFAC 3.5 public. The resulting tool AliBaba2 available http://wwwiti.cs.uni-magdeburg.de/grabe/alibaba2.