Search

Introduction In recent years, a number of software tools for data quality control and instrument checks have been described, including both targeted and untargeted proteomics, with or without using isotopic labeling, and standard samples. The goal of this project was to create a web-based application to enable users of any skill level to generate performance metrics using commercial protein digest standards. By utilizing metadata and extensive training data, there is no need to fine tune parameters, thereby allowing users with little experience to evaluate instrument performance and guide troubleshooting. Methods We mimicked common technical issues encountered in the lab occurring anywhere between the peptide mixture and the m/z signal to build or repurpose existing models to evaluate the performance of the autosampler, chromatography (including column health and mobile phase chemistry), ionization, selection and fragmentation. Using a commercially available HeLa digest, 20 runs were collected with varying data acquisition and peptide loads in addition to a longitudinal dataset of 14 runs collected over one year on the same instrument with constant conditions. The analyses were fully automated and a web application developed using Shiny, allowing upload and analysis using instrument-specific spectral libraries generated from the commercial HeLa digest and common contaminants. Preliminary Data Whereas most QC pipelines can generate chromatograms and report numbers of identified peptides, ratios or absolute abundances of selected peptides across runs, our web application creates a single, visual, high-level overview of the entire LC-MS system, helping identify where something went wrong, if it did, and guide the instrument troubleshooting. This includes, but is not limited to, estimating the volume and amount injected, the mobile phase composition and pH, the column condition, contaminants and background signals, ionization, precursor selection and fragmentation. These models, and metrics derived from them, contextualize the reported numbers of identified spectra, peptides and proteins. In this process, all available data (MS1, MS2, retention times) and metadata is used. Basic parameters such as resolving power and mass measurement accuracy are also reported. Practically important, the web application can be used from any instrument control computer with a web browser and internet connection - no software installation is necessary. For longitudinal comparison within one laboratory, a stand-alone version is also available. Ongoing work includes gathering interlaboratory data and increasing the coverage of commercially or community-developed proteomics standards, including E. coli and yeast digests, as well as peptide standards. Novel Aspect Unlike comparable tools, our QC Benchmarker supports commercially available standards and allows easy identification of potential instrument issues.