Comparison of Fast Scanning Data Dependent and Data Independent Acquisition Methods for a Multi-OMIC Cancer Study Using High-Speed Chromatography

Significance of the Topic

Metabolomic and lipidomic analyses enable detailed investigation of molecular alterations associated with disease, treatment, and lifestyle factors. In colorectal cancer research, rapid high-throughput methods help capture dynamic biochemical changes in plasma samples, supporting biomarker discovery and improved clinical diagnostics.

Objectives and Study Overview

The study aims to evaluate and compare data-dependent acquisition (DDA) and data-independent acquisition (DIA) modes on a high-speed chromatography platform coupled to a Xevo MRT mass spectrometer. Plasma from healthy controls and colorectal cancer patients (colon and rectum) was profiled to identify distinguishing metabolic and lipid markers.

Methodology

• Sample preparation involved protein precipitation of human plasma using chilled isopropanol.
• Chromatographic separations were performed on an ACQUITY Premier UPLC system with reversed-phase columns and a 5-minute gradient.
• Mass spectrometry acquisitions included:

• DDA at 20 Hz MS and 50 Hz MS/MS.
• Full-scan MSE DIA at 20 Hz.
• Stepped quadrupole DIA (SONAR) to generate cleaner, DDA-like fragmentation spectra.

Instrumentation Used

• Waters ACQUITY Premier UPLC for high-speed liquid chromatography
• Waters Xevo MRT mass spectrometer for high-resolution accurate-mass detection
• Data Convert for generic file format conversion (mzML)
• Skyline and MS-DIAL for data processing, peak picking, and compound identification
• Metacore for pathway enrichment analysis

Main Results and Discussion

• Principal component analysis of DIA data demonstrated tight clustering of replicates and clear separation between healthy controls and colorectal cancer groups, confirming high reproducibility (CV ≤ 20%).
• Stepped quadrupole DIA generated spectra with enhanced specificity and cleaner fragmentation patterns, resulting in identification scores comparable to DDA.
• Lipid profiling revealed significant under-expression of lysophosphatidylcholines (LPCs) and up-regulation of ceramide species in cancer patients.
• Database searching (±1 ppm tolerance) against in-house libraries and LipidMaps provided high-confidence lipid assignments.
• Pathway analysis highlighted dysregulation of the WNT signaling cascade, aligning with known colorectal carcinogenesis mechanisms.

Benefits and Practical Applications

• High throughput: 5-minute LC gradients improved sample throughput without compromising data quality.
• Flexibility: Both DDA and DIA modes demonstrated robust performance on the same instrument platform.
• Accuracy: Sub-1 ppm mass accuracy for precursors and fragments ensured reliable identifications.
• Compatibility: Vendor-neutral workflows and open file formats facilitated third-party software integration.

Future Trends and Potential Uses

• Further refinement of quadrupole DIA methods to deepen proteome and lipidome coverage.
• Integration of machine learning for automated feature selection and identification.
• Standardization of high-throughput omics workflows for clinical diagnostics and personalized medicine.
• Cloud-based informatic platforms to streamline data sharing and collaborative research.

Conclusion

The comparison of DDA and DIA strategies on a high-speed chromatography platform demonstrated that both approaches deliver high-quality, reproducible multi-omic data. Stepped quadrupole DIA combines the specificity of DDA with the throughput of full-scan methods, providing an effective workflow for colorectal cancer biomarker discovery.

References

• Tsugawa H et al. MS-DIAL: data-independent MS/MS deconvolution for comprehensive metabolome analysis. Nat Methods 12, 523–526 (2015).
• Nguyen L et al. Pathways of Colorectal Carcinogenesis. Gastroenterology 158, 291–302 (2019).