DIGITAL BIOLOGY
Features
Radical Numerics' phage therapy was a joint effort with Tabula; George Church resigns from a scientific advisory role
Radical Numerics' Oct. 20 announcement of AI-designed phages given to a patient did not mention Tabula Bioscience. The New York Times reports that Tabula did much of the work, including making the phage cocktail; both companies say each designed some of the phages. Radical CEO Eric Nguyen said their contract required keeping the seven-month collaboration confidential. George Church, who also advises Tabula, resigned from a scientific advisory role at Radical on Nov. 5, rejecting that explanation. Tabula says it led the therapy.
The White House will require federally funded labs to buy synthetic DNA only from certified suppliers
A Nov. 5 memorandum from national security adviser Marco Rubio and science adviser Michael Kratsios makes compliance a condition of federal life sciences funding. Suppliers must run every order of 50 or more nucleotides past a NIST-verified screener, verify customers and keep records for eight years; benchtop synthesizers need a screener's authorization to run. NIST has 90 days to issue standards. Funded labs must switch to certified suppliers within 250 days, 340 for benchtop equipment. Violators can lose funding and be barred for up to five years. It replaces the 2024 screening framework.
Isomorphic Labs is in talks to raise about $5 billion at a roughly $50 billion valuation
Bloomberg reported the talks on Nov. 4, citing people familiar with the matter; the financing has not closed and terms could change. A month earlier, Bloomberg put the talks at a valuation of at least $40 billion. Isomorphic's programs are still preclinical. Its president, Max Jaderberg, said on Nov. 2 that the company is aiming for first clinical trials by the end of the year, R&D World reported.
AutoDNA, an AI-run lab, carried nucleic acid work from synthesis to sequencing across 11,000 steps
The system, from Shanghai Jiao Tong University researchers, orchestrates more than 20 instrument types; humans only maintain hardware and restock reagents and labware. It matched expert-optimized enzymatic DNA synthesis yields, ran concurrent protocols 3.2 times faster, completed a DNA data storage cycle with exact recovery and validated a clinically detected KRAS G13D mutation. The code is open source on GitHub. Nature ran a commentary and a news story alongside the paper.
Chai Discovery says Roche will use its AI models in a number of drug discovery programs
The release says the programs cut across therapeutic areas but names none and gives no financial terms. It follows Chai's October deal with GSK. Chai describes its latest model, Chai-3, as achieving high target success rates and binding affinities, including to challenging targets, but publishes no figures.