Content

Because most proteins do not work alone, but rather associate with other proteins to perform their functions, it is important to know about these associations. It does not help that these associations can take different forms, such as physical contact in protein-protein interactions, participation in complex formation, or as links in the same signalling or metabolic pathway. To elucidate as many such protein associations as possible, without being restricted to binary interactions, we therefore proposed a novel approach to detect protein association based on the re-processing of large-scale, public mass spectrometry based proteomics data. This approach is based on inferring protein association based on the co-occurrence of proteins across many different proteomics experiments.

Concretely, we reprocess data from the PRIDE repository using our in-house ReSpin pipeline (which is in turn built from our pride-asap , SearchGUI and PeptideShaker tools) to identify all proteins in each experiment. We then use the peptide counts for each protein in each experiment to determine highly co-occurring proteins across various mass spectrometry based experiments. See the reference below for details on the underlying method.

The online Tabloid Proteome is the collection of these co-occurring protein pairs, together with their existing biological relation in existing knowledgebases. In addition to the protein associations we derived, pathway links from Reactome, protein-protein interactions from IntAct and BioGRID, protein complexes from CORUM, and paralog information from Ensembl are also superimposed. Functional annotation is provided by disease information from DisGeNET, and Gene Ontology annotations.

You can search that online Tabloid Proteome in a variety of intuitive ways; just have a look at home page!