Related Experiment Video
Updated: Sep 4, 2025
09:19
NMR-Based Fragment Screening in a Minimum Sample but Maximum Automation Mode
Published on: June 4, 2021
3.4K
Machine Learning-Based Fragment Selection Improves the Performance of Qualitative PRM Assays.
Patrick M Vanderboom1, Santosh Renuse1,2, Anthony D Maus1
1Department of Laboratory Medicine and Pathology, Division of Clinical Biochemistry and Immunology, Mayo Clinic, Rochester, Minnesota 55905, United States.
Journal of Proteome Research
|July 18, 2022
Summary
Machine learning enhances targeted mass spectrometry assays for detecting SARS-CoV-2. This approach improves sensitivity in qualitative PRM assays compared to traditional methods, aiding in more accurate sample discrimination.
Area of Science:
- Biomarker Discovery
- Proteomics
- Machine Learning Applications
Background:
- Targeted mass spectrometry is crucial for protein biomarker detection.
- Traditional assays require manual selection and set limits (LOD, LLOQ) sacrificing sensitivity for specificity.
- Existing methods face limitations in balancing sensitivity and specificity.
Purpose of the Study:
- To investigate the application of machine learning in qualitative parallel reaction monitoring (PRM) assays.
- To enhance the discrimination of positive from negative samples in PRM assays.
- To compare the performance of machine learning models against traditional methods for PRM assay analysis.
Main Methods:
- An ensemble machine learning model was trained using targeted PRM data from nasopharyngeal swabs.
- The dataset included 282 SARS-CoV-2 positive and 994 negative samples.
- The model was validated on an independent set of 200 positive and 150 negative samples.
Main Results:
- The machine learning model achieved 92% sensitivity compared to RT-PCR.
- A traditional approach using the same data yielded 86.5% sensitivity.
- Machine learning demonstrated superior performance in discriminating positive from negative samples.
Conclusions:
- Machine learning can be effectively applied to qualitative PRM assays.
- ML-enhanced PRM assays offer superior sensitivity and specificity over traditional methods.
- This approach holds promise for improved biomarker detection in clinical settings.
