#25 · Proteomics

Prior-knowledge networks coupled to proteomic statistics for phosphosite interpretation

The common thread is coupling statistical proteomic or phosphoproteomic outputs with prior-knowledge biomolecular networks, delivered as software. MSstatsBioNet, a Bioconductor package, links MSstats differential protein abundance results to INDRA literature-derived and curated networks. A separate preprint extends DrDimont multi-omics networks with kinase interactions and applies them to pulmonary arterial hypertension microvascular endothelial cells, highlighting MAPK signalling. A third item applies nested linear regression to pan-cancer proteomics of 1,006 tumours, linking 3,038 phosphosites to interacting-protein abundance and yielding a prioritized resource. What is new is the integration of network priors with statistical outputs in package or platform form, rather than standalone analyses. The visible summaries report no benchmark comparisons, so performance remains untested.