DIGGER 2.0: digging into the functional impact of differential splicing on human and mouse disorders

Abstract

Changes in alternative splicing between groups or conditions contribute to protein-protein interaction rewiring, a consequence often neglected in data analysis. The web server and database DIGGER overcomes this limitation by augmenting a protein-protein interaction network with domain-domain interactions and splicing information. Here, we present DIGGER 2.0, which now features both experimental and newly added predicted domain-domain interactions. In addition to the human interactome, DIGGER 2.0 adds support for mouse as an important model organism. Additionally, we integrated the splicing analysis tool NEASE, which allows users to perform online splicing- and interactome-informed enrichment analysis on RNA-seq data. In two application cases (multiple sclerosis and mice models of cardiac diseases), we show the utility of DIGGER 2.0 for deeper exploration and functional interpretation of changes in alternative splicing in human and mouse disorders. DIGGER 2.0 is available at https://exbio.wzw.tum.de/digger/.

Graphical Abstract

Figure 1.

Overview of the NEASE workflow. (A) The input data include differentially used exon skipping events as a list or output from MAJIQ, rMATS, or Whippet. (B) The user selects parameters of the NEASE analysis. (C) DIGGER 2.0 performs the analysis of affected protein features. (D) DIGGER 2.0 performs the splicing-aware pathway enrichment based on the chosen database (e.g. KEGG, Reactome). (E) DIGGER 2.0 results include the list of enriched pathways. (F) KEGG and Reactome pathways can be visualized as clusters. (G) The user can analyze an individual pathway.

Figure 2.

Most significant enrichment results sorted by NEASE score; (A) without predicted interactions; (B) using high- and medium-confidence predicted interactions. Overall, both methods have a high agreement, but some pathways have gained more importance (lower P-values) when using pDDIs.