Abstract P327: COPa Library: A Proteomic Knowledge Base for Cardiovascular Biology and Medicine

Abstract

We constructed the Cardiac Organellar Peptide Atlas Library (COPa library) as a targeted and interactive resource to the cardiovascular community. Annotated peptide spectra are hosted using a relational database in a modular fashion based on species (e.g. human, mouse) and organelles (e.g. mitochondria, proteasome). Within this release of COPa library, a total of 108,268 spectra have been disseminated via two avenues. A web portal was established to navigate the library via parallel set of identifiers, such as protein name, accession number, gene symbol, etc. In parallel, a web-service cyber-infrastructure was engineered to aid the annotation of mass spectra submitted via internet. The large raw spectra files are dissected into small data packages at the local PC before submission. This workflow surpasses the limitation of network bandwidth, as well as enables parallel data submission and search. A benchmark test with 897,327 ms/ms spectra showed the library searching covers 93.4% of proteins identified via database searching, as well as additional 23.9% of proteins at the same level of statistical confidence. In addition, a wiki-like web interface was embedded in the library web portal in order to facilitate the synthesis of consensus knowledge among the cardiovascular community on innovations of functional proteomics. Overall, the COPa library search supports targeted proteomic characterization, which complements database search for exploratory survey. The implementation of the COPa library-based proteomic knowledgebase leverages state-of-the-art technology and annotated datasets among the research community at large. Its application bridges discovery-driven and hypothesis-driven research while fostering translational medicine.