Benchmarking A Visual Acuity Ion Classifier for MS/MS Deconvolution and 4 Identification of Native Membrane Proteins In Vitro

Significance of the Topic

The ability to accurately deconvolute complex tandem mass spectra is critical for advancing native and top-down mass spectrometry workflows, particularly when studying membrane proteins and large protein assemblies. Parameter-free and robust algorithms reduce user bias and accelerate data interpretation, supporting applications in structural biology, biopharmaceutical characterization, and quality control.

Objectives and Study Overview

This work evaluates the performance of a visual acuity ion classifier within ExDViewer 4.6 for automated, parameter-free deconvolution of MS/MS spectra. The study benchmarks its accuracy and sensitivity on diverse peptide and protein samples, including challenging intact membrane proteins and proteoliposomes, across multiple mass spectrometer platforms.

Methodology and Instrumentation

Samples comprised targeted peptides, model proteins (carbonic anhydrase, monoclonal antibody), and three proteoliposomes containing VAMP2, Aquaporin Z, and semiSWEET. Data were acquired by:

• Direct infusion and liquid chromatography–MS/MS on an Agilent 6545XT AdvanceBio LC/Q-TOF with an ExD cell for electron-based fragmentation.
• Native top-down analyses on a Thermo Q Exactive UHMR upgraded with a Legacy ExD cell.
• Targeted TDMS on an Agilent 6545XT Q-TOF equipped with an ECD fragmentation cell.

ExDViewer 4.6 freeware was used with default settings to assign IonScore values (0–15) to isotopic clusters, considering profile peak shape, centroid m/z error, hydrogen transfer events, and neighboring peak overlap.

Main Results and Discussion

Benchmarking with over 10,000 expert-annotated fragment ions demonstrated that a scoring threshold of 1.5 captured more than 90% of detectable ions while maintaining under 2% incorrect assignments. In native TDMS of in vitro membrane proteins isolated from E. coli polar lipid extracts, the method achieved 44% sequence coverage for VAMP2, 70% for semiSWEET, and 30% for Aquaporin Z. The deconvolution algorithm sensitively detected low-intensity species and enabled rapid manual validation without extensive false positives.

Benefits and Practical Applications

The parameter-free design allows novice users to analyze complex top-down spectra as readily as bottom-up data, without adjusting default parameters. Advanced users retain the option to fine-tune settings. Fast, accurate deconvolution supports high-throughput workflows in structural proteomics, biopharmaceutical QA/QC, and membrane protein research. Custom amino acids and post-translational modifications can be integrated into the target editor, extending the tool’s applicability.

Future Trends and Applications

Ongoing efforts will compare ExDViewer’s performance to other deconvolution tools across broader user groups and diverse sample types. Future developments may include machine learning-driven peak assignment, expanded support for hybrid fragmentation methods, and enhanced integration with laboratory information management systems (LIMS). The capability to process non-standard amino acids and novel modifications opens pathways for studying engineered proteins and complex biomolecules.

Conclusion

ExDViewer 4.6 demonstrates robust, parameter-free deconvolution of MS/MS spectra, delivering high sensitivity and accuracy for both small peptides and large membrane protein complexes. Its intuitive interface accelerates manual validation, making advanced top-down analysis accessible to a wide range of users.

References

1. Panda A, Giska F, Duncan AL, et al. Direct determination of oligomeric organization of integral membrane proteins and lipids from intact customizable bilayer. Nat Methods. 2023;20(6):891–897.
2. Guthals AL, Sturgeon D, Gavrilenko A, et al. Parameter-free deconvolution and visualization of peptide and protein fragmentation mass spectra. ASMS 2023.
3. ExDViewer 4.6 Freeware. Agilent Technologies. Available at exdviewer.agilent.com.