Fig. 1: The integrated workflow for discovery and identification of SEPs in Y. pestis.

It includes three modules: prediction, discovery, and curation. In the prediction module, the RNA-seq data are processed and assembled to generate a searchable SEP database. In the discovery module, the bacteria samples were prepared by a refined peptidomics approach, respectively, with eight proteases, followed by LC-MS analysis. The MS raw data were searched by PEAKS engine against the predicted SEP database with a known protein database as search background. In the curation module, the resulting SEPome was processed by BLASTP to remove the known proteins and then filtered at the level of peptide-spectrum match (PSM), and then the N terminus was determined to generate the final SEPome. Refer to Supplementary Data 1 and 2 for the details of SEP candidates. MWCO molecular weight cut-off, UP unique peptide, AA amino acid, SAAV single amino acid variant.