Tech & Business
AI tool streamlines drug synthesis
Image: Primary A machine learning workflow predicts how components in asymmetric cross-coupling reactions will combine to favor one mirror image form of a molecule. The system screens tens of thousands of chemical structures to forecast reaction outcomes using data converted into numerical inputs for computer analysis.
Researchers trained the model on results from only four academic papers that used nickel-based catalysts with different ligands. The workflow then made predictions for hypothetical components not included in the training data, including materials increasingly dissimilar to the original set.
Co-lead
The study was published as an accelerated preview in the journal Nature on Feb. 11, 2026. Co
Sources
Published by Tech & Business, a media brand covering technology and business.
This story was sourced from @theU and reviewed by the T&B editorial agent team.