Automation and machine learning in schistosomiasis drug discovery

Artificial intelligence (AI) and machine learning (ML) are rapidly transforming pharmaceutical research, providing new capabilities for drug target identification, compound screening, lead optimization, and clinical trial design. Despite growing industry investment, AI-originated drug candidates remain a small fraction of clinical pipelines, and late-stage approvals remain limited, underscoring the need for rigorous biological validation alongside computational innovation. Schistosomiasis remains a major neglected tropical disease (NTD), with transmission reported in 79 countries and at least 253.7 million people requiring preventive chemotherapy in 2024, approximately 93.9% of whom live in Africa. Praziquantel, developed in the 1970s, remains the sole widely deployed therapy. Limited commercial incentives have historically suppressed research and development (R&D) investment in NTDs, creating a structural gap that AI and automation are well-positioned to address. This review synthesizes current knowledge on how automation and ML can accelerate schistosomiasis drug discovery. We describe principal biological assays for assessing parasite viability, examine AI-driven strategies across the full drug development pipeline from target identification to regulatory submission and evaluate computational tools validated against schistosomes. We further assess AI-driven drug repurposing as a cost-effective near-term strategy and critically discuss limitations, including data fragmentation and heterogeneity, gaps in assay standardization, and model interpretability. Our central argument is that meaningful progress requires not only algorithmic innovation, but equal investment in standardized, high-throughput phenotypic platforms capable of generating data-rich training inputs for ML systems.