P186-T Evaluation of 36 MudPIT Replicate Datasets Using Normalized Spectral Counting and PLGEM
Year: 2007
Type: article
Abstract: Multidimensional protein identification technology (MudPIT) is a platform integrating multidimensional liquid chromatography and tandem mass spectrometry for protein profiling. As MudPIT becomes increasingly popular, the full characterization of large datasets remains challenging. A total of 36 MudPIT technical replicates from two biological samples were acquired on XPs and LTQs ion-trap tandem mass spectrometers. This study evaluated the applicability of label-free quantitative proteomic information by fitting mean and standard deviation of normalized spectral abundance factors (NSAFs) to a power law global error model (PLGEM). Protein fractions from affinity purifications of 293 cells expressing various Flag-tagged human Mediator subunits were enzymatically digested and analyzed by MudPIT. A total of 475 unique proteins were identified from 36 runs, resulting in a 36 × 475 matrix of relative abundance (NSAF values). We observed a “goodness of fit” when PLGEM was applied to the entire 36 × 475 dataset, and subsets of the data containing only NSAF values from XP or from LTQ, indicating that NSAF datasets are compatible with this modeling scheme. Specifically, log-log plots of standard deviation vs. mean NSAFs revealed a linear dependence, with slopes ranging from 0.66 to 0.75 and an adjusted r(2) equaled to 0.99. In addition, the residuals were found to be normally distributed across the entire dynamic range of average NSAF values. Similar ranges and distribution were also observed in Affymetrix GeneChip datasets. This study demonstrates the utilities of NSAFs and PLGEM in statistically analyzing MudPIT datasets. With additional modeling using datasets from more complex samples or with “spiked in” experiments, this approach has a potential to become an integral data-mining tool for proteomics datasets acquired from one mass spectrometer as well as across multiple instrumentation platforms.
Source: PubMed Central
Language: English
Subfield: Spectroscopy
Field: Chemistry
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W25166609