At the core of many biological experiments is the goal of comprehensively answering the question: What are the effects of a perturbation? Perturbations are any functional alteration in a biological system and their genome- wide effects can be measured by innumerable sequencing technologies. Nascent transcription assays, like GRO- seq, PRO-seq, NET-seq, and TT-seq, are emerging methods to measure the direct effect of a perturbation on transcription. To better leverage genome-wide nascent RNA sequencing data to more fully understand the effects of perturbations on nascent transcription, new computational methods for identifying regions differentially affected by experimental and control conditions must be developed and applied. I propose to address this need by developing PFinder, a computational tool capable of identifying the regions affected by a perturbation. I will demonstrate its utility by analyzing NET-seq data from a reverse genetic screen in S. cerevisiae to provide insight into the roles of transcription regulatory proteins, in terms of both location and scale of effect. This will create an annotation of the yeast genome with the transcriptional consequences of knocking out each factor, lending new insight into transcription regulation. I will next apply PFinder to other nascent RNA sequencing data collected from several human cell types treated with small molecules to uncover their novel effects. The results from this proposal will establish a method that can be applied to any nascent RNA sequencing data to identify loci affected by a perturbation genome-wide and without a prior hypothesis. PFinder and the insights it provides will enable more thorough exploitation of nascent RNA sequencing data, both previously published and yet to be generated.
Most scientific experiments ask variations of the same fundamental question: What are the effects of a perturbation? Nascent transcription assays are emerging methods to measure the direct effect of a perturbation on transcription, but analysis is often focused by a hypothesis; unanticipated effects, therefore, can easily be missed. In this proposal, I will create a bioinformatics tool, PFinder, that will identify regions across the genome transcriptionally affected by a perturbation and apply it to multiple sequencing data types from multiple organisms.
