Leave-One-Out-Cross-Validation Based Machine Learning Algorithms for Efficient, Scalable, and Robust Prediction of Artemisinin Resistance in Plasmodium falciparum Kelch13 mutants.

A leave-one-out cross-validation machine-learning framework using 40 Plasmodium falciparum Kelch13 variants and 10 physicochemical/structural descriptors enables rapid prediction of ART-sensitive versus ART-resistant phenotypes.

• Why it matters: Existing Kelch13-based phenotype assignment is too time-intensive for real-time surveillance, and the ring-survival assay is lab-adaptation dependent and may not match clinical ART responsiveness.
• What they did: The approach computes each variant’s relative deviation from wild-type K13 across ten descriptors, forming a compact feature set to train and validate an AI classifier with leave-one-out cross-validation.
• The result: The model is designed to screen more than 100 K13 variants within hours, supporting near-real-time triage of newly emerging Kelch13 mutations for ART-S/R prediction.

AI summary of the abstract · may contain errors · not medical advice · read the paper.