Final Report Summary - PTRCODE (Systematic study of post-transcriptional regulation mediated by RNA-binding proteins and miRNAs) Post-transcriptional regulation (PTR) is a pervasive and complex system controlling every step of RNA metabolism including stability, decay, localization and translation. PTR mechanisms depend on the ability of RNA-binding proteins (RBPs) and microRNAs (miRNAs) to interact with their target sites on mRNAs. Recent studies show that RBPs and miRNAs can promote or antagonize each other’s effects. However, the majority of the literature so far has studied the regulatory effect RBPs and miRNAs individually. In this project, we aim to consider the regulatory effects of RBPs, miRNAs and their interactions concurrently. Such systematic studies have been difficult due to the scarcity of knowledge on RBP and miRNA binding sites. With the recent explosion of high-throughput techniques, experimental characterization of binding sites of hundreds of RBPs and miRNAs are now available for multiple cell lines. We aim to develop computational models that utilize these datasets to achieve the following aims: (i) genome-wide mapping of RBP and miRNA binding sites together with the RNA secondary structure and conservation information, (ii) identification of cooperative and competitive interactions between RBPs and miRNAs, (iii) integrative analysis of RBPs, miRNAs and the interactions between them to understand post-transcriptional regulatory events, (iv) combining transcriptional and post-transcriptional regulatory networks to explain differential gene expression in cancer.We succeeded in compiling a comprehensive mapping of RBP and miRNA binding sites on mRNAs. To this end, we incorporated RNA secondary structure predictions, CLIP datasets, conservation and binding motifs. We utilized this genome-wide map to identify cooperative and competitive interactions between RBPs and miRNAs. We also showed that these interactions could have effects on PTR by analyzing the genome-wide expression changes upon the knockdown of a set of well-characterized RBPs. These analyses confirmed that competition with other factors weakens the regulatory activity of RBP, and cooperation has the opposite effect. Using counts of RBP and miRNA binding sites in 3’UTRs, we were able to predict stability and steady-state expression of transcripts in a number of cell types. We also investigated the roles of PTR in cancer. We identified several differentially expressed RBPs in distinct cancer types. We used copy-number variation, DNA methylation, and regulatory effects of TFs, miRNAs and RBPs as features within a lasso-regularized regression framework to predict gene expression in cancer. We applied this model to lung squamous cell carcinoma (LUSC) and liver hepatocellular carcinoma (HCC). We identified the most significant features using a feature selection procedure and found that some of these factors are predictive of survival in cancer. In the last part of the project, we updated our genome-wide binding site compendium with recently available eCLIP datasets. We also considered the effect of alternative polyadenylation (APA) by quantifying the number of binding sites in proximal and distal 3’UTR isoforms separately. Using counts of binding sites that incorporate APA effects resulted in a better performance in predicting gene expression in cancer. Lastly, we also investigated potential sponge relationships between lncRNAs and RBPs. For this purpose, we mapped the binding sites of RBPs on lncRNAs and performed several analyses that compare the expression of lncRNAs, RBPs and the target genes of RBPs. All this information is available on a public website for other researchers.In summary, our compendium of binding sites of RBPs and miRNAs have been instrumental in (i) identifying the interactions between RBPs and miRNAs, (ii) predicting mRNA expression and stability (iii) predicting gene expression in cancer, (iv) Identifying potential decoy lncRNAs for RBPs. Results of these analyses confirmed the importance and the complexity of post-transcriptional regulation in gene expression control in both normal and cancer cell lines. Additionally we utilized the compiled data sets and binding sites to successfully collaborate with several other research groups. Project website: http://hkazanlab.com/ptr_research.html
