Darvesh Gorhe

I am a PhD candidate in the lab of Marko Jovanovic working on modeling crosslinking and immunoprecipitation sequencing (CLIP-seq) data using stochastic processes and mixture models.

CLIP-seq gives us the RNA sequences bound by RNA-binding proteins (RBPs). Many labs have developed CLIP-seq variants that improve sensitivity (iCLIP, eCLIP), specificity (easyCLIP, PAR-CLIP), and throughput (SPIDR, ABC). Analysis methods like Skipper, CIMS, Piranha, PureCLIP, omniCLIP, etc aim to parse this data to tell us: which RBPs are binding to which genes more than background. However, I would argue these methods don’t yield much information unique to CLIP-seq data. We often find ourselves using other methods like cryo-EM to validate the claims from CLIP-seq data. My work aims to address this gap by looking at distribution of the crosslink spacing rather than the concentration of crosslinking events within a window.