Inter-session reproducibility measures for high-throughput data sources

Area of Science:

• Biotechnology
• Clinical Diagnostics
• Biomarker Discovery

Background:

• High-throughput biological assays (e.g., microarrays, mass spectrometry) show promise for clinical disease detection and biomarker evaluation.
• Current research often overlooks data variability arising from multiple data-generation sessions, a critical factor for clinical implementation.

Purpose of the Study:

• To develop and validate a methodology for assessing the reproducibility of high-throughput data across multiple data-generation sessions.
• To address the challenge of multi-session effects in the application of high-throughput assays to clinical practice.

Main Methods:

• Proposed a novel framework for measuring and testing reproducibility across different data-generation sessions.
• Applied and demonstrated the framework using mass spectrometry (MS) data from four distinct sessions using identical samples.

Main Results:

• The developed framework successfully measured and tested reproducibility aspects of high-throughput data across multiple sessions.
• Demonstrated the framework's utility on real-world mass spectrometry data, highlighting its practical applicability.

Conclusions:

• Assessing multi-session effects is critical for the reliable clinical application of high-throughput biological assays.
• The proposed methodology provides a robust approach to quantify and ensure data reproducibility in clinical settings, facilitating biomarker discovery and disease detection.