MhcVizPipe: A Quality Control Software for Rapid Assessment of Small- to Large-Scale Immunopeptidome Datasets

Abstract

MS-based immunopeptidomics is maturing into an automatized and high-throughput technology, producing small- to large-scale datasets of clinically relevant major histocompatibility complex (MHC) class I-associated and class II-associated peptides. Consequently, the development of quality control (QC) and quality assurance systems capable of detecting sample and/or measurement issues is important for instrument operators and scientists in charge of downstream data interpretation. Here, we created MhcVizPipe (MVP), a semiautomated QC software tool that enables rapid and simultaneous assessment of multiple MHC class I and II immunopeptidomic datasets generated by MS, including datasets generated from large sample cohorts. In essence, MVP provides a rapid and consolidated view of sample quality, composition, and MHC specificity to greatly accelerate the "pass-fail" QC decision-making process toward data interpretation. MVP parallelizes the use of well-established immunopeptidomic algorithms (NetMHCpan, NetMHCIIpan, and GibbsCluster) and rapidly generates organized and easy-to-understand reports in HTML format. The reports are fully portable and can be viewed on any computer with a modern web browser. MVP is intuitive to use and will find utility in any specialized immunopeptidomic laboratory and proteomics core facility that provides immunopeptidomic services to the community.

Keywords: MHC; MS; immunopeptidomics; peptide; software.

Graphical abstract

Figure 1

Overview and installation of the MhcVizPipe (MVP) software.A, illustration showing the (i) experimental workflow for the isolation and identification of MHC-associated peptides by LC–MS/MS, (ii) the software tools integrated into the MVP pipeline for rapid data processing, and (iii) the HTML report generated to support (or not) downstream immunopeptidomic data interpretation through quality control (QC) and quality assurance (QA). The MVP pipeline was created with the Python language and parallelizes in-house (UpsetPlotly and Plotly-Logo) and established (NetMHCpan, NetMHCIIpan, and GibbsCluster) algorithms. B, installation steps. The user can download and install the MVP software from http://github.com/CaronLab/MhcVizPipe on Linux, Mac, and Windows 10 (using the Windows Subsystem for Linux). The installation requires the download of the third-party software tools and the installation of MVP.

Figure 2

Overview of the MhcVizPipe (MVP) graphical user interface (GUI). To run an analysis, the user can copy–paste a list of peptides or upload more than one file at a time. Samples can be labeled with detailed information. The user click the “GO!” button to start an analysis, and a loading screen appears while the analysis is running, followed by a pop-up window with a link to the HTML report. More details about the GUI are available at https://github.com/CaronLab/MhcVizPipe/wiki/Usage. A command line interface (CLI) is also available for “batch” analyses at https://github.com/CaronLab/MhcVizPipe/wiki/Command-line-interface.

Figure 3

Visualization of the QC scores (LF and BF) is calculated inTable 1.A and B, histograms and heatmaps illustrating high-quality (A) and middle- to low-quality (B) immunopeptidomic data generated by MS from various tumor biopsies (Table 1). Histograms showing the distribution of peptides according to their length (right panel) and predicted MHC binding affinity (left panel). Heatmaps are automatically generated by MVP and provided in the HTML report. For each HLA allele, NetMHCpan provides a %Rank score for individual peptides, which are color coded on the heatmap in red, blue, and yellow for SB, WB, and NB, respectively, with a linear gradient between the colors. BF, binding fraction; HLA, human leukocyte antigen; LF, length fraction; MHC, major histocompatibility complex; MVP, MhcVizPipe; NB, nonbinder; QC, quality control; SB, strong binder; WB, weak binder.

Figure 4

Speed performance of the MVP software for small- to large-scale QC analyses.A, Pie charts showing the time estimates to process and analyze the selected datasets manually (human-based in green) or using MVP (computer-based in red). The following datasets were tested (from left to right): one QC sample (JY), one PBMC sample, 15 PBMC samples, five mouse tissue samples, and 20 mouse tissue samples. B, large-scale QC analysis of the HLA peptidome of 152 samples. Graph showing the total number of peptides and the two scores calculated based on peptide lengths (LF) and the number of peptides predicted to bind HLAs (BF) (y-axis). Each sample has three points, one for each of the values (shown in the legend). The points for each sample are connected by a vertical line. C, zoom in on the sample “BCN_13-Plasma_A.” Histogram and heat map showing the length distribution and predicted HLA-binding affinity (%Rank) of the detected peptides in this sample. LF and BF scores are indicated. BF, binding fraction; HLA, human leukocyte antigen; LF, length fraction; MVP, MhcVizPipe; PBMC, peripheral blood mononuclear cell; QC, quality control.