김지현 (울산대학교 의과학과)

김지현 (울산대학교 의과학과)

Mass spectrometry-based proteomics, combined with machine learning, offers powerful capabilities for molecular diagnosis and subtype classification in complex diseases. In this study, we present an integrative approach using high-resolution mass spectrometry and computational modeling to identify membrane protein biomarkers from peripheral blood mononuclear cells (PBMCs) for the diagnosis and stratification of Acute Myeloid Leukemia (AML). Membrane proteins were enriched from PBMCs of AML patients and healthy donors, enabling surface proteome profiling. Quantitative proteomic data were subjected to feature selection and then used to train predictive models with supervised machine learning algorithms, including logistic regression, support vector machines, and random forest classifiers. Distinct membrane protein signatures were identified, uncovering AML-relevant functional pathways. Patient samples were stratified into subtypes based on the French-American-British (FAB) classification, with high-risk subtypes like M4 and M5 distinguished. This study demonstrates the potential of surface proteomics for biomarker discovery and shows how machine learning can translate proteomic data into diagnostic and prognostic insights, advancing precision medicine in hematologic oncology. This work was supported by the National Research Foundation of Korea (NRF), funded by the Korean government (MSIT) (No. RS-2024-00454407).