Protein inference is a central issue in proteomics, given the presence of shared peptides (i.e., peptides that might originate from different proteins sharing homology, from different proteoforms due to alternative mRNA splicing, post-translational modifications, proteolytic cleavages, and/or allelic variants). Indeed, in bottom-up mass spectrometry-based proteomics, the most widely used proteomic approach, peptide-protein connectivity is lost for experimental reasons and protein identifications are to be inferred from peptide identifications. Shared peptides can generate quite complex peptide-to-protein mapping structures but these can be efficiently represented using bipartite graphs, with peptides and proteins as vertices and with edges featuring peptide to protein membership. Graph connected components (CCs) (i.e., the largest subgraphs in which any two vertices are connected to each other by a path and not connected to any other of the vertices in the supergraph) can be used as a mesure of the level of ambiguity in protein identifications. Proteins sharing one or more peptides are gathered in the same CC (multi-protein CCs), while unambiguous protein identifications are represented by CCs with a single protein vertex (single-protein CCs). CCs represent a peptide-centric strategy to group proteins, independent from the variety of protein-centric strategies of protein grouping and protein inference. As such, it does not require protein inference and it is widely applicable, reproducible and transparent.

The CCs4prot package allows to build a graph from shotgun proteomic identifications and calculate its connected components.

The availability of an increasing number of sample-matched proteomic and transcriptomic datasets can be exploited to reduce ambiguity of protein identifications. Indeed, according to the central dogma of biology, there can be no protein without the corresponding transcript. Following this, protein identifications for which the corresponding transcript is identified in the sample-matched transcriptome are more likely to be correct than those with no expressed transcript.

The CCs4prot package implements a transcriptome-informed filtering strategy to reduce ambiguity of protein identifications and allows to measure the impact of the filtering on ambiguity by assessment of the proportion and size of multi-protein CCs and by visual inspection of peptide-to-protein mappings for ambiguous identifications.

git clone https://github.com/laurafancello/CCs4prot.git

library("devtools")
devtools::install("CCs4prot")