Evaluating cPILOT Data toward Quality Control Implementation

Abstract

Multiplexing enables the monitoring of hundreds to thousands of proteins in quantitative proteomics analyses and increases sample throughput. In most mass-spectrometry-based proteomics workflows, multiplexing is achieved by labeling biological samples with heavy isotopes via precursor isotopic labeling or isobaric tagging. Enhanced multiplexing strategies, such as combined precursor isotopic labeling and isobaric tagging (cPILOT), combine multiple technologies to afford an even higher sample throughput. Critical to enhanced multiplexing analyses is ensuring that analytical performance is optimal and that missingness of sample channels is minimized. Automation of sample preparation steps and use of quality control (QC) metrics can be incorporated into multiplexing analyses and reduce the likelihood of missing information, thus maximizing the amount of usable quantitative data. Here, we implemented QC metrics previously developed in our laboratory to evaluate a 36-plex cPILOT experiment that encompassed 144 mouse samples of various tissue types, time points, genotypes, and biological replicates. The evaluation focuses on the use of a sample pool generated from all samples in the experiment to monitor the daily instrument performance and to provide a means for data normalization across sample batches. Our results show that tracking QC metrics enabled the quantification of ∼7000 proteins in each sample batch, of which ∼70% had minimal missing values across up to 36 sample channels. Implementation of QC metrics for future cPILOT studies as well as other enhanced multiplexing strategies will help yield high-quality data sets.

Establishing Quality Control Metrics for Large-Scale Plasma Proteomic Sample Preparation

“…However, specifications for TMT labeling exceeded and trended well with previously reported multiplexing experiments. 22,39,40 LC−MS/ MS data acquisition methods used are recommended to more closely resemble final analysis methods 32 for the overall cohort study. For fractionation, BSA can be added to automated digestion plate layouts to decrease the operator workload.…”

“…External controls are added as interbatch checks that are used to assess instrument performance. 38,39 Together, internal and external QC metrics strengthen confidence in the LC−MS/MS analysis and study data, in addition to enabling other researchers to reproduce and build upon published studies. 21 This study sought to establish QC metrics for sample preparation steps toward the analysis of a large cohort (N = 808) using an automated, high-throughput plasma proteomic workflow.…”

“…Previously, our laboratory established analytical metrics to assess QC in LC−MS/MS 40 and LC−MS 3 data acquisition for tandem mass tag (TMT)-based proteomics analysis. 39 Those metrics easily allowed operators to determine when a sample batch was out-of-specification (OOS) and enabled real-time troubleshooting. Here, we created five QC sample types (QC std , QC dig , QC pool , QC TMT , and QC BSA ) and used them strategically to (1) measure instrument performance, (2) check the efficiency of the individual workflow steps, and (3) guide operator actions that progressed the sample preparation workflow.…”

“…Internal controls can be added as a QC check of multiple sample preparation steps, LC–MS acquisition, , and data normalization. External controls are added as interbatch checks that are used to assess instrument performance. , Together, internal and external QC metrics strengthen confidence in the LC–MS/MS analysis and study data, in addition to enabling other researchers to reproduce and build upon published studies …”

“…This study sought to establish QC metrics for sample preparation steps toward the analysis of a large cohort ( N = 808) using an automated, high-throughput plasma proteomic workflow. Previously, our laboratory established analytical metrics to assess QC in LC–MS/MS and LC–MS data acquisition for tandem mass tag (TMT)-based proteomics analysis . Those metrics easily allowed operators to determine when a sample batch was out-of-specification (OOS) and enabled real-time troubleshooting.…”