Virtual screening dataset for predicting multistage antimalarial activity using ML-QSAR models

This dataset contains the results of a large-scale virtual screening performed using machine learning QSAR (ML-QSAR) models (Borba et al., 2024) trained to predict multistage antimalarial activity.The dataset includes 1,025,012 small molecules from commercially available compound libraries, with predicted probabilities of activity against Plasmodium asexual blood stages (3D7 and W2 strains), P. berghei ookinete, liver stage, and gametocytes.Each entry includes the compound identifiers (InChIKey and SMILES), molecular weight, and predicted probabilities for each parasite stage, along with aggregated scores for compound prioritization.This dataset supports a manuscript currently under peer review