Talus Bioscience Inc, a Seattle-based drug discovery company, has received a USD 1.03 million SBIR R44 grant from the National Center for Advancing Translational Sciences (NCATS) to advance an AI-driven platform targeting transcription factors — a protein class long considered inaccessible to small-molecule drugs.
Transcription factors regulate gene expression across virtually every disease-relevant cellular process, yet their structural complexity — including disordered regions and large protein complexes — has frustrated conventional drug discovery. Talus is building around this problem with two proprietary technologies: TF-Scan, a high-throughput functional proteomics platform that measures compound effects on transcription factor activity across thousands of targets in live cells, and Strategian, an AI platform that uses machine learning models trained on TF-Scan data to predict how small molecules interact with transcription factor-chromatin complexes. The grant will fund optimization of protein and compound representations within Strategian, active learning cycles to identify inhibitors for three undisclosed high-priority transcription factors, and development of cloud-based data infrastructure — including a SaaS web application for internal programs and external partners.
The R44 mechanism, which supports late-stage small business innovation research with commercial intent, signals NCATS confidence in the platform's translational potential. Principal investigator William E. Fondrie leads the project. The transcription factor targeting space has attracted growing industry interest, with companies including Syros Pharmaceuticals and Relay Therapeutics pursuing related approaches through distinct methods; Talus differentiates through its functional proteomics readout rather than structural or chemogenomic screening alone.
This article was generated with AI assistance and reviewed and edited by the AllSci editorial team Explore more at AllSci News: https://allsci.com/news/
Spot something wrong? Report an issue with this article
