Cross-platform Clinical Proteomics using the Charité Open Standard for Plasma Proteomics (OSPP)
Description
we present the Charité Open Peptide Standard for plasma proteomics (OSPP), an open resource composed of 211 isotope-labeled peptides, intended to be used as an internal standard for plasma and serum proteomic projects. The OSPP was designed based on peptides consistently quantified across a panel of diverse human studies, is made of peptides that can easily be synthesized, that distribute equally over chromatographic gradients, and that show consistent identification performance across diverse LC-MS platforms, acquisition methods, and matrices.
Files
Steps to reproduce
Discovery proteomics The raw proteomics data from all DIA methods was processed using DIA‐NN, 1.8.1, available on GitHub (DIA‐NN GitHub repository 67). The MS2 and MS1 mass accuracies were set to 20 and 12 ppm (ZenoTOF 7600 data) or 15 and 15 ppm (timsTOF and Exploris 480 data), and the scan window to 7. The aforementioned OSPP-specific Human Spectral Library is used for data processing with additional commands: --fixed-mod SILAC,0.0,KR,label --channels SILAC,L,KR,0:0; SILAC,H,KR,8.014199:10.008269 --peak-translation --original-mods --matrix-ch-qvalue 0.01 --restrict-fr --report-lib-info Specifically, following a two-step MBR approach 46, an in silico spectral library is first generated by DIA-NN from the FASTA file(s); this library is then refined based on the DIA dataset and subsequently used to reanalyze the dataset, to obtain the final results. The data were filtered in the following way. First, a 1% run-specific q-value filter per isotope channel was automatically applied at the precursor level by DIA-NN (--matrix-ch-qvalue 0.01). We note that in any experiment processed using the MBR mode in DIA-NN, 1% global precursor q-value filtering is also applied automatically 46. For quantification, we used “Precursor.Translated” value as quantities for each precursor in MS2 quantification. For Exploris 480 data, since orbitraps are sensitive in MS1, we also used “Ms1.Translated” was used. Targeted proteomics Zeno MRM-HR data were processed using Skyline (64-bit, v.23.1.0.268). No blinding was performed during peak integration. The quantity of each peptide is calculated by the summation of peak areas of each selected fragment of a peptide (list of fragments used for quantification in Supplement Table 4). Calibration curve The calibration curve for each of the 211 peptides was either accepted or rejected based on a set of rules and criteria: the limit of quantification (-LLOQ and –ULOQ) was determined based on the accuracy of replicated injection on the same LC-MS platform (Supplementary Table 6). Peptide concentration (expressed in pg/µl) was determined from calibration curves, constructed with native and isotopic labeled peptide standards in the surrogate matrix (4 ng/µl BSA), and manually inspected and validated. Peptides with > 40% of values below the lowest or above the highest detected calibrant concentration across all samples were removed from the analysis. Linear regression analysis of each calibration curve was performed using custom R code (with 1/x weighting).