Detecting cooperatively bound transcription factor pairs on the genome from high throughput data

Vishaka Datta, NCBS Bangalore

The transcription of many eukaryotic genes is regulated by the binding of multiple transcription factors (TFs) within short distances of each other. In such regulatory regions, the binding of one TF can increase, decrease, or leave unchanged the affinity of an adjacent TF for its target sequence. These affinity altering interactions have been well documented in in vitro experiments, but the extent to which the binding of a given TF in vivo affects a second TF bound nearby is unclear. The influence of these interactions can be understood through data from high throughput sequencing techniques such as ChIP-seq and ChIP-exo. These methods give a list of genomic locations bound by a pair of TFs in vivo, as well as a (noisy) measure of their binding affinities at each location. We propose a method that can find those locations where the a pair of TFs interacted while binding DNA. The highlight of this method is that it uses only the binding affinities of each TF for detecting interactions while not taking the underlying sequence that is bound into consideration.