Cancer is a disease of the genome: sequential accumulation of DNA mutations lead to both direct and indirect changes in the structure and abundance of the proteins that perform most of the functional activities within the cell. While methodologies for the analysis of cancer genomes and transcriptomes have undergone rapid benchmarking and standardization, our understanding of how best to analyze the cancer proteome remains less-developed. In particular, there are key questions remaining in how to infer the abundances of peptides not detected in a subset of samples, in optimizing database searches to detect cancer-specific peptides caused by point-mutations, alternative transcript isoforms or fusion genes, in understanding the association between DNA, mRNA and protein data, and in performing robust absolute quantitation.
To address these issues, we have set up the NCI-CPTAC DREAM Challenge as a community-based collaborative competition of researchers from across the world working together to answer key questions in cancer proteomics, focused around the integration of diverse data-types. The CPTAC effort includes the collection of tumour/normal pairs with matched genomic, transcriptomic and proteomic characterization, which provide a unique data to run the proteogenomic challenge.