Improved functional annotation of posttranslationally modified proteins

Identification of posttranslational modifications (PTMs) by mass spectrometry is biased towards abundant proteins, skewing our understanding of PTMs towards readily detected proteins. To uncover specifically enriched Gene Ontology (GO) terms associated with PTMs, we developed a method to account for protein abundance bias when performing GO term enrichment analyses on PTM datasets. GO term enrichment analysis is frequently used to examine “-omics” datasets for overrepresented functional terms within a subset of the dataset, such as regulated genes or modified proteins. However, the identification of PTMs is biased to abundant proteins that are more readily detected in the mass spectrometer. Therefore, GO enrichment analyses comparing posttranslationally modified proteins (referred to here as modified proteins) to unmodified proteins are likely to reveal overrepresented GO terms associated with abundant proteins, rather than modified proteins specifically. The method we developed here provides more accurate GO term enrichment analysis of PTMs and identifies terms whose enrichment results from the bias to identify PTMs on abundant proteins.