Microsoft's RetroChimera retrosynthesis model published in Nature, open-sourced under MIT license
Microsoft Research · rss · 2026-09-21
Microsoft Research published RetroChimera, a retrosynthesis prediction model, in Nature, and open-sourced its implementation and weights (MIT license, also on Microsoft Foundry).
- Architecture: combines two complementary sub-models — Transformer-based R-SMILES 2 generates precursors de novo (flexible but prone to hallucination), while GNN-based NeuralLoc selects reaction templates (more grounded but limited by template coverage)
- Ensembling: a learned learning-to-rank strategy votes across both models' predictions, handling both common and rare reaction classes
- Results: in blind tests, PhD-level chemists preferred RetroChimera's single-step disconnections over prior models and even literature-recorded reactions; it succeeded on 9 of 10 challenging multistep targets versus 5/4/2 for baselines
- Impact: could accelerate design-make-test cycles in drug discovery and materials, enabling closed-loop self-improving synthesis planning with lab automation
More from Apps
- Amazon blocks Muse shopping agent as Shopify embraces agentic checkout — rickasaurus · 2026-09-22
- Why your AI prompts get worse every time you patch them — alex_verem · 2026-09-22
- WIRED: Meta's Muse Agent Excels at Data Collection More Than Getting Tasks Done — mmitchell_ai · 2026-09-22
- Underdog local AI assistant praised as best consumer experience since ChatGPT — CCgong · 2026-09-22
- China's agent wars: Xiaomi ships OS-level agents, Tencent blocks them, Doubao shops on Douyin — bookwormengr · 2026-09-22
- How to claim your monthly Google Cloud credits: $10 for AI Pro, $100 for Ultra — gaganghotra_ · 2026-09-22