Mining ChIP data for evidence of mechanisms underlying transcription factor DNA-occupancy. Qiong Cheng¹, Majid Kazemian¹, Hannah Pham², Charles Blatti¹, Michael Brodsky², Saurabh Sinha¹. 1) Department of Computer Science, UIUC, IL 61801; 2) Department of Molecular Medicine, University of Massachusetts Medical School, MA 01655.

   ChIP-based genome-wide assays of TF occupancy have emerged as a powerful, high throughput method to understand transcriptional regulation. This has led to great interest in the underlying biochemical mechanisms that direct TF-DNA binding, with the ultimate goal of computationally predicting a TFs occupancy profile in any cellular condition. In this study, we examined the influence of various potential determinants of TF-DNA binding on a much larger scale than previously undertaken. We used a thermodynamics-based model of TF-DNA binding, called STAP, to analyze 45 TF-ChIP data sets from Drosophila embryonic development. We built a cross-validation framework to use a baseline model that takes into account only the ChIPed (primary) TFs motif and compare it with more complex models where binding by secondary TFs are hypothesized to influence the primary TFs binding. Candidate interacting TFs were chosen based on RNA-SEQ expression data from the time point of the ChIP experiment. We found widespread evidence of both cooperative and antagonistic effects by secondary TFs. We were able to identify multiple classes of interactions, including long-range interactions between primary and secondary motifs, suggestive of indirect effects such as chromatin remodeling, short-range interactions with specific inter-site spacing biases, suggestive of direct physical interactions, and overlapping binding sites suggesting competitive binding. We found evidence that the TFs ZELDA and TRL may synergistically promote the binding of several primary TFs in early and mid-stage embryonic development respectively, while EXD, RETN, JIGR1 and TTK can antagonistically influence TF-DNA binding in many cases, relying on both accessibility-mediated and accessibility-independent mechanisms. Finally, we conducted in vitro pull-down assays to test model-based predictions of short-range cooperative interactions, and found that seven of the nine TF pairs tested physically interacted.