Radical Numerics Launches Omnii-1 Biology Model

Radical Numerics announced in a press release the launch of Omnii-1, a genome language model that learns from DNA, RNA, proteins and other molecular data. The company says researchers used the model to design a bacteriophage that was administered to a patient with a drug resistant bacterial infection.

The researchers designed phages targeting the patient's infection within weeks. A cocktail containing one of the AI designed phages killed the patient's bacterial isolate in laboratory testing before a physician administered it through compassionate use access.

Omnii-1 converts 28 types of biological data, including epigenetic signals, into a common representation. Radical Numerics is demonstrating the model across eight applications, including cancer vaccine design, analysis of genetic causes for rare diseases and biological threat detection.

The company says Omnii-1 identified AI-generated pathogen sequences that conventional screening tools missed during testing. Radical Numerics is also collaborating with Lawrence Livermore National Laboratory on biological threat detection.