Comparative analysis of mRNA degradation and protein degradation in prostate tissues indicates high stability of proteins
To benchmark the PIN algorithm that quantifies the extent of protein degradation in samples, we generated a set of “ground truth” samples, in which the levels of proteome degradation were known and independently validated by western blotting. Protein extracts of HeLa Kyoto cells were treated with the low specificity protease Proteinase K at different protease concentrations.
Sample Processing Protocol
Six samples of protein extracts were treated with sequentially increasing amounts of Proteinase K ranging from 0.0005 μg/μl to 0.02 μg/μl, while maintaining a constant substrate concentration of 1 μg/μl. All treated samples were incubated at 25℃ for five minutes and were rapidly transferred to heat at 100℃ for five minutes. To best simulate the practical situation of a clinical sample cohort and to set up the null distribution for outlier detection in the follow-up statistical test, nine samples were prepared as controls without treatment of Proteinase K. All samples were then subjected to complete tryptic digestion prior to mass spectrometric analysis. Each sample, including the controls, were prepared in two biological replicates, and were then acquired in the same way as described in the main text on a TripleTOF 5600 mass spectrometer operated in SWATH-MS mode. Two technical replicates were measured per sample. In addition, 28 shotgun MS measurements, two biological replicates and two technical replicates from seven samples (six treated samples plus one control sample), were acquired for SWATH assay library construction. In total, the benchmarking dataset consists of 60 SWATH-MS injections and a SWATH assay library constructed from 28 shotgun injections.
Data Processing Protocol
The SWATH .wiff files were first converted into profile mzXML using msconvert. Through the iPortal workflow manager, the resulting 60 SWATH-MS mzXML files were analyzed by OpenSWATH with default settings. After the targeted extraction of fragment ion chromatograms, pyprophet was used to calculate a single discriminant score from a subset of the scores (library_corr yseries_score xcorr_coelution_weighted massdev_score norm_rt_score library_rmsd bseries_score intensity_score xcorr_coelution log_sn_score isotope_overlap_score massdev_score_weighted xcorr_shape_weighted isotope_correlation_score xcorr_shape) and to estimate the q-value to facilitate FDR control. TRIC was then run on the pyprophet results to perform the feature alignment to re-rank peak groups obtained in the original targeted extraction stage with the following parameters (realign_method: spline, dscore_cutoff: 1, target_fdr: 0.01, max_rt_diff: auto_3medianstdev, method: global_best_overall).
Shao W, Guo T, Toussaint NC, Xue P, Wagner U, Li L, Charmpi K, Zhu Y, Wu J, Buljan M, Sun R, Rutishauser D, Hermanns T, Fankhauser CD, Poyet C, Ljubicic J, Rupp N, Rüschoff JH, Zhong Q, Beyer A, Ji J, Collins BC, Liu Y, Rätsch G, Wild PJ, Aebersold R. Comparative analysis of mRNA and protein degradation in prostate tissues indicates high stability of proteins. Nat Commun. 2019 10(1):2524 PubMed: 31175306