Computational chemists reported machine-learning and free-energy-guided design of new tuberculosis KasA inhibitors. The study, published in Molecular ... , describes a fragment-based recombination strategy that generates novel chemical entities for KasA, a tuberculosis drug target. The approach focuses on reducing the hit-finding burden by using physics-based simulations alongside ML to prioritize compounds before synthesis. In an era of rising drug resistance, new early-stage chemical matter for TB targets remains a priority for both academia and biotech. While the news is preclinical, it adds to the emerging pipeline picture for TB drug discovery where structure-informed computational methods are increasingly used to generate candidates with better binding potential.