Abstract
Predicting transcription factor binding sites (TFBS) from sequence is one of the most challenging problems in computational biology. The development of (semi-)automated computer-assisted prediction methods are needed to find TFBS over an entire genome, which is a first step in reconstructing mechanisms that controls gene activity. Bioinformatics journals continue to publish diverse methods for predicting TFBS on a monthly basis. To help practitioners in deciding which method to use to predict for a particular TFBS, we provide a platform to assess the quality and applicability of the available methods. Assessment tools allow researchers to determine how methods can be expected to perform on specific organisms or on specific transcription factor families. This chapter introduces the TFBS detection problem and reviews current strategies for evaluating algorithm effectiveness. In this chapter, a novel and robust assessment tool, the Motif Tool Assessment Platform (MTAP) is introduced and discussed.