Grant: Project Grant

Determining how gene regulation is orchestrated is the next frontier in human genetics and genomics. A key element in this problem is knowing how DNA-binding regulatory proteins known as "transcription factors" recognize specific DNA sequences, in order to perform the regulation. This knowledge is needed to understand the functional impact of sequence variations that contribute to human diversity, disease susceptibility, and dysregulaton in cancer. A major hurdle is that the sequence specificity of many supposed DNA binding proteins is either completely unknown or poorly described. We have developed powerful laboratory assays and associated computational expertise which can address this problem, as well as an industry-standard database to make the information publicly available. In a recent study, we found that transcription factors often have much higher specificity than thought. Our study was aimed at the completely uncharacterized transcription factors, and it was successful in this regard, but one of the main outcomes was that previously-studied transcription factors - which we incorporated as controls - also often have very high specificity. Here, we propose to apply these systems to complete the catalog of human transcription factor-DNA sequence specificity by applying our methods to the >1,000 previously-studied transcription factors. We will also dissect the mechanisms that confer the unanticipated specificity, modify our databases to accommodate the new information and possibly new computational models, and explore possible functions of the transcription factors with highest specificity.