New preprint on trusworthy ML for anti-cancer drug sensitivity prediction

Big news! Our preprint “Increasing trustworthiness of machine learning-based drug sensitivity prediction with a multivariate random forest approach” is now live on ChemRxiv! We introduce MORGOTH, a trustworthy multivariate random forest model that simultaneously performs classification and regression. Additionally, MORGOTH provides a graph representation of the random forest to address model interpretability, and a cluster analysis of the leaves to measure the dissimilarity of new inputs from the training data to account for its reliability and robustness. While broadly applicable, we focus on anti-cancer drug sensitivity prediction and prioritization. We show that it has state-of-the-art performance while being more trustworhty than any other existing approach.
For more information check out our paper.