Statistical control of peptide and protein error rates in large-scale targeted data-independent acquisition analyses

Abstract

Liquid chromatography coupled to tandem mass spectrometry is the main method for high-throughput identification and quantification of peptides and inferred proteins. Within this field, data-independent acquisition (DIA) combined with peptide-centric scoring, exemplified by SWATH-MS, emerged as a scalable method to achieve deep and consistent proteome coverage across large-scale datasets. Here we discuss the adaptation of statistical concepts developed for discovery proteomics based on spectrum-centric scoring to large-scale DIA experiments analyzed with peptide-centric scoring strategies and provide guidance on their application. We show that optimal tradeoffs between sensitivity and specificity require careful considerations of the relationship between proteins in the samples and proteins represented in the spectral library. We propose the application of a global analyte constraint to prevent accumulation of false positives across large-scale datasets. Furthermore, to increase the quality and reproducibility of published proteomic results, well-established confidence criteria should be reported for detected peptide queries, peptides and inferred proteins.

The Generation of a Comprehensive Spectral Library for the Analysis of the Guinea Pig Proteome by SWATH‐MS

“…However, the stochastic nature of parent ion selection can introduce variability to peptide identification outputs, hinder quantification between sample runs, and thus necessitate lengthy and costly procedures such as sample fractionation (to reduce input complexity) and injection replicates . The adoption of data‐independent acquisition (DIA) MS modes such as sequential window acquisition of all theoretical mass spectra (SWATH‐MS), whereby MS2 fragment ion spectra are collected for each parent ion observed, in a series of mass‐to‐charge isolation windows, has presented the opportunity to overcome these issues and obtain deep, label‐free proteomic coverage of complex samples in a timely manner without fractionation . Interpretation of SWATH spectra, however, requires reference to a spectral library of peptide sequence matching (including established m / z and LC retention time co‐ordinates), itself oft‐obtained from the outcomes of multiple DDA runs.…”

Combining Rapid Data Independent Acquisition and CRISPR Gene Deletion for Studying Potential Protein Functions: A Case of HMGN1

“…To analyse the DIA‐MS data, we used both spectrum‐centric (named directDIA in Spectronaut, a method extracting pseudo‐ spectrum directly from DIA data without the need of an assay library) and peptide‐centric (based on SWATH assay library which contains mass spectrometric assays for 10 000 human proteins, named panHuman hereafter) approaches. Both approaches achieved substantial sensitivity when both peptide‐ and protein‐ FDR were strictly controlled at 1% (Figure A,B) . On average, directDIA identified and quantified 5 465 ± 7 Swiss‐Prot proteins, assigned from 61 476 ± 430 unique peptides (≈82 840 ± 685 peptide precursors) in the 12 samples, whereas panHuman library detected 5625 ± 27 proteins corresponding to 57 920 ± 1601 unique peptides (≈68 338 ± 2092 peptide precursors).…”

“…SWATH‐MS has proven to be reproducible in a cross‐lab study for >4000 proteins quantified in mammalian cells . Recently, the statistical strategies controlling the quality of protein identification and quantification in large‐scale DIA‐MS have matured . Because proteins catalyze and control directly most biological processes, and because proteotype characterization essentially bridges the genotypic variation and phenotype diversity, we propose that a sensitive, reproducible, robust, and rapid proteomic measurement enabled by, for example, DIA‐MS would provide a general and powerful alternative approach to study potential protein functions when combined with a CRISPR experiment, to not only verify the successful knockout of the target gene (and its protein expression), but also to profile perturbed signaling pathways and quantitative phenotypes.…”

Systems proteomics approaches to study bacterial pathogens: application to Mycobacterium tuberculosis

“…To extract quantitative data from the highly complex and convoluted fragment ion spectra generated by DIA/SWATH-MS, prior knowledge about the fragmentation properties of specific peptides is typically employed [35]. During the past years, custom computational tools have been developed enabling chromatogram extraction [36,37], false discovery rate control [38], protein quantification [39], differential expression analysis [40,41] and study of post-translational modification for SWATH-MS [42]. …”