Search

Abstract As part of the advancement in therapeutic decision-making for brain tumor patients at St. Jude Children’s Research Hospital (SJCRH), we developed three robust classifiers, a deep learning neural network (NN), k-nearest neighbor (kNN), and random forest (RF), trained on a reference series DNA-methylation profiles to classify central nervous system (CNS) tumor types. The models’ performance was rigorously validated against 2054 samples from two independent cohorts. In addition to classic metrics of model performance, we compared the robustness of the three models to reduced tumor purity, a critical consideration in the clinical utility of such classifiers. Our findings revealed that the NN model exhibited the highest accuracy and maintained a balance between precision and recall. The NN model was the most resistant to drops in performance associated with a reduction in tumor purity, showing good performance until the purity fell below 50%. Through rigorous validation, our study emphasizes the potential of DNA-methylation-based deep learning methods to improve precision medicine for brain tumor classification in the clinical setting.

Abstract In this study, we leveraged machine-learning tools by evaluating expression of genes of pharmacological relevance to standard-AML chemotherapy (ara-C/daunorubicin/etoposide) in a discovery-cohort of pediatric AML patients (N = 163; NCT00136084 ) and defined a 5-gene-drug resistance score (ADE-RS5) that was predictive of outcome (high MRD1 positivity p = 0.013; lower EFS p