Towards Automated Quality Assessment and Analysis of Targeted Mass Spectrometry Data using Machine Learning

Shadi Toghi Eshghi, Paul Auger, Stephen Schauer, Kristin Wildsmith, W. Rodney Mathews
OMNI-Biomarker Development, Genentech Inc., South San Francisco, CA

Warning: Undefined variable $headshot in /var/www/html/view_abstract/view_abstract_in_program.php on line 704

Shadi Eshghi (Presenter) Genenech

Presenter Bio: Shadi Eshghi is a scientist in the biomarker development department at Genentech. Her work focuses on development of bioinformatics tools and techniques to facilitate analysis and interpretation of mass spectrometry, flow cytometry and mass cytometry data. Prior to joining Genentech, Shadi worked in Dr. Hui Zhang’s lab at Johns Hopkins University on novel computational and experimental methods for exploring the glycome and glycoproteome using mass spectrometry. Shadi obtained a Ph.D. in biomedical engineering from Johns Hopkins University and a B.Sc. in electrical engineering and is a 2016 Siebel Scholar.

Relevant Financial Disclosures (within past 24 months) Salary Genentech Discussing Products? NO, presenter will NOT present educational content related to products/services of the ineligible company(ies) listed above. Conflict mitigation NOT REQUIRED.

Targeted mass spectrometry has enabled highly multiplexed and sensitive quantitation of lipids, metabolites and protein biomarkers for clinical research and development. The increased rate of data production along with the need for improving data processing efficiency and reproducibility has created growing demand for automated analytical tools in the field. We have developed TargetedMSQC, an open source R package that takes advantage of machine learning for semi-automated quality assessment of chromatographic peaks in targeted MS data. The workflow of TargetedMSQC and its application for analysis and quality control of large datasets including clinical trial biomarker data will be discussed.