RAPDOR: Using Jensen-Shannon Distance for the computational analysis of complex proteomics datasets

Abstract

The computational analysis of large proteomics datasets, such as those from gradient profiling or spatially resolved proteomics, is often as crucial as the experimental design. We present RAPDOR, a tool for intuitive analyzing and visualizing such datasets, based on the Jensen-Shannon distance and subsequent analysis of similarities between replicates, applied to three datasets. First, we examined the in-gradient distribution profiles of protein complexes with or without RNase treatment (GradR) to identify the set of RNA-binding proteins (RBPs) in the cyanobacterium Synechocystis sp. PCC 6803. RBPs play pivotal regulatory and structural roles; although numerous RBPs have been identified, the complete set is unknown for any species. RAPDOR identified 80 potential RBPs, including ribosomal proteins, likely RNA-modifying enzymes, and several proteins not previously associated with RNA binding. High-ranking putative RBPs, such as the universal stress protein Sll1388, or the translation inhibitor LrtA/RaiA, were predicted by RAPDOR but not the TriPepSVM algorithm, indicating uncharacterized RBP domains. These data are available online at https://synecho-rapdor.biologie.uni-freiburg.de, providing a comprehensive resource for RNase-sensitive protein complexes in cyanobacteria. We then show by reanalyzing existing datasets, that RAPDOR is effective in examining the intracellular redistribution of proteins under stress conditions. RAPDOR is a generic, non-parametric tool for the intuitive and versatile analysis of highly complex data sets such as the study of protein distributions using fractionation protocols.