JUMPg: an Integrative Proteogenomics Pipeline Identifying Unannotated Proteins in Human Brain and Cancer Cells

ABSTRACT: Proteogenomics is an emerging approach to improve gene annotation and interpretation of proteomics data. Here we present JUMPg, an integrative proteogenomics pipeline including customized database construction, tag-based database search, peptide-spectrum match filtering, and data visualization. JUMPg creates multiple databases of DNA polymorphisms, mutations, splice junctions, partially trypticity, as well as protein fragments translated from the whole transcriptome in all six frames after RNA-seq de novo assembly. We use a multistage strategy to search these databases sequentially, in which the performance is optimized by re-searching only unmatched high quality spectra, and re-using amino acid tags generated by the JUMP search engine. The identified peptides/proteins are displayed with gene loci using the UCSC genome browser. The JUMPg is applied to process a label-free mass spectrometry dataset of Alzheimer’s disease postmortem brain, uncovering 496 new peptides of amino acid substitutions, alternative splicing, frame shift, and “non-coding gene” translation. The novel protein PNMA6BL specifically expressed in the brain is highlighted. We also tested JUMPg to analyze a stable-isotope labeled dataset of multiple myeloma cells, revealing 991 sample-specific peptides that include protein sequences in the immunoglobulin light chain variable region. Thus, the JUMPg program is an effective proteogenomics tool for multi-omics data integration.

INSTRUMENT(S): Q Exactive

ORGANISM(S): Homo Sapiens (human)

TISSUE(S): Brain

DISEASE(S): Alzheimer's Disease

SUBMITTER: xusheng wang  

LAB HEAD: Junmin Peng

PROVIDER: PXD004010 | Pride | 2017-02-20

REPOSITORIES: Pride