Translating Quantitative Proteomics Data into Biological Insight: Methodology for Replicate Proteomics Data Analysis in the Context of Type 2 Diabetes

摘要

1. We have implemented a prototype algorithm that solves the following: 2. Input: Replicated quan ta ve data from COMPASS, including protein and phospho-isoforms from a high-throughput experiment and a list of comparisons 3. Output: list of signi fi cant measurement-comparison pairs. 4. We have included transcript and microarray quan ta on, allowing us to compare protein with transcript, phospho-isoform with protein, imputed mo fs, and imputed kinases. 5. We effi ciently perform 39 biologically meaningful comparisons on 89,584 measurements, for a total of 3.5 million measurement-comparisons. 6. We found 101,402 signifi cant measurement-comparisons (3% of possible) (10% FDR) 7. We found that 9,314 measurements changed signifi cantly in some comparison (10% of possible) (10% FDR). 8. We have started verifying biologically relevant results discovered using this data.