Peptide Analysis for Complex Communities
Pacific Northwest National Laboratory performs non-genome specific peptide analysis to compare changes in ecosystem protein profiles across environmental conditions.
Identification of putative peptides expressed by periphyton in response to 238U uptake. Enlarged Detailed View
Current global proteomic strategies are critically dependent on the existence of a genome annotation database for each organism under study. This currently limits uncovering changes in ecosystem proteins by high-throughput protein identifications and potential biomarker discovery to systems where the organisms have compiled genome annotations. Our method circumvents this critical limitation by estimating putative peptide profiles, regardless of their initial identifications. The ecosystem-level, putative peptide profiles are obtained with current proteomic identification algorithms, such as Sequest, using a non-redundant protein database containing proteins from sequenced species related to the ecosystem of study to match experimental and virtual/database generated spectra. A suite of internal standards are also employed for sample- to-sample and run-to-run normalization, and to monitor quality. REML-fitted mixed-effects statistical modeling are utilized and offer a flexible, well-established methodology for comparative proteomics studies integrating common experimental designs with LC-MS sample processing plans. Putative peptides of interest are then sequenced using de novo techniques to facilitate further characterization of the peptide. Thus, a genome-specific annotation, unattainable for almost all wild ecosystem communities, would not be needed to compare changes in the ecosystem protein profile across conditions.
Publication
Daly DS, KK Anderson, EA Panisko, SO Purvine, R Fang, ME Monroe and SE Baker. 2008. "A Mixed Effects Statistical Model for Comparative LC-MS Proteomics Studies." Journal of Proteome Research, EPub Feb 6, 7(3):1209-17.
Contacts: Ellen Panisko, Scott Baker