Label-Free Protein-RNA Interactome Analysis Identifies Khsrp Signaling Downstream of the p38/Mk2 Kinase Complex as a Critical Modulator of Cell Cycle Progression
Fig 2
Inference of outlying RBP-client interactions from changes in client transcripts.
(A) Using gene expression levels, numbers of altered client mRNAs were plotted against number of known client mRNAs for each respective RBP. A linear correlation could be identified between the number of changed client mRNAs and known client mRNAs. (B) Studentized residuals (outlyingness), leverage (potential to influence the linear model) and influence analysis (represented by the size to point) are represented through influence plots. Data points perturbing the model were identified by high leverage and studentized residuals. Outliers representing RBPs with higher number of changed client mRNAs were identified through high absolute values of standardized residuals. The same was done by (C) plotting number of upregulated clients against number of changed clients, as well as using vector information on (D) differential promoter usage, (E) differential splicing, and (F) differential CDS. DNA damage-related RBPs—Elavl1, Tia1, Tial1, Srsf1, Srsf2 could be identified through RBP-client analysis.