Donor-wise metabolomic prediction of platelet storage duration: a proof-of-concept study
Related Experiment Video
Updated: Nov 4, 2026
LC-MS Analysis of Human Platelets as a Platform for Studying Mitochondrial Metabolism
Published on: April 4, 2016
1Department of Blood Transfusion, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei, China.
Background:
Whether the metabolomic profile of stored platelets can predict storage duration across donors under donor-wise validation is unknown. We assessed a multi-task learning framework.
Materials And Methods:
Utilizing the public metabolomic dataset ST003937 (239 samples, 24 donors, 111 metabolites), a dual-task machine learning framework was constructed to predict storage duration (Task A) and biotinylation status (Task B). Models were evaluated by donor-wise 5-fold cross-validation with random-split comparison. Inter-donor distanceswere quantified and SHAP (SHapley Additive exPlanations) identified keymetabolites. Multi-task learning was also performed.
Results:
Task A failed to generalize, with a mean area under the curve (AUC)of 0.46, improving to 0.59 under random split, whereas Task B was robust(AUC 0.77). Median inter-donor distance (6.35) far exceeded the storage groupdistance (1.73). Multi-task learning marginally improved Task A by 0.03. SHAPand Gini importance identified metabolites related to energy metabolism andoxidative stress as key features for Task B.
Discussion:
Inter-donor metabolic variability exceeded storage-inducedchanges in magnitude, explaining why cross-donor prediction of storageduration was unsuccessful despite documented group-level metabolicdifferences. The robust AUC gap between validation strategies confirmsa "donor fingerprint" effect. Donor-specific baselines may be useful forfuture assessment of platelet quality. Multi-task learning provided marginalimprovement. Task B's robust performance validates the analytical framework.
