Identification of Low Abundance Proteins in a Highly Complex Protein Mixture

This is a summary of the 2014-2016 multi-lab study conducted by the ABRF Proteomics Research Group. 

Proteome Profiling of whole cell lysates is a proteomic service performed by most, if not all, core facilities as a routine service. However, capture of low-level proteins remains challenging. Sample preparation, overall fractionation time as well as instrument sensitivity are main factors that determine the outcome of such an experiment. The ABRF Proteomics Research Group
(PRG) conducted an inter-laboratory study to assess current capabilities in identification of low abundance proteins in a highly complex protein sample among core facilities. The PRG provided participants with lyophilized HeLa cell lysates with four proteins spiked –in at three different amounts (20 fmol, 100fmol and 500 fmol) . Participants were expected to perform protein identification using sample preparation and LC methods of their choice to detect these four spiked-in proteins, which had with no homology to human proteins. Participants did not know identities of the proteins, rather, the PRG provided a database that contains de-identified protein sequences for these four proteins. For each participating laboratory, we report the overall number
of proteins identified, the number of spectra and unique peptides identified for spiked-in in proteins at the different amounts. Preliminary results indicate that participating laboratories were able to identify all four spiked-in proteins at 500 fmol. As expected, the identification of all four spiked-in proteins varied in different laboratories at the lower levels of spiked-in amounts. We found a significant advantage of performing fractionation on complex samples to  detect proteins at an extremely low amount. The utility of fractionation on detecting spiked-in proteins at higher amounts appears to be protein-dependent.