Radical Numerics

Systems, scaling and architecture for general biological intelligence

[Radical Numerics @RadicalNumerics]

1/ At Radical Numerics, we have a dual mandate: advancing biological design and biodefense.

Omnii Defense is our latest advancement in using genome language models for biological threat detection, including previously unseen and engineered threats.

Powered by Omnii-1, the system can run inference on biological sequences in real time, even directly from a portable sequencer connected to a laptop, all done locally and securely.

Our vision is to make advanced biological threat detection accessible anywhere, from research labs to the front line.

Watch our new 5-minute film on Omnii Defense, and read the exclusive story in @WIRED.

[Radical Numerics @RadicalNumerics]

@WIRED 12/ We’re seeking partners to evaluate and deploy Omnii Defense across cloud and field settings. Explore the results and get in touch: https://t.co/0YE6AT17gX

[Radical Numerics @RadicalNumerics]

1/ At Radical Numerics, we have a dual mandate: advancing biological design and biodefense.

Omnii Defense is our latest advancement in using genome language models for biological threat detection, including previously unseen and engineered threats.

Powered by Omnii-1, the system can run inference on biological sequences in real time, even directly from a portable sequencer connected to a laptop, all done locally and securely.

Our vision is to make advanced biological threat detection accessible anywhere, from research labs to the front line.

Watch our new 5-minute film on Omnii Defense, and read the exclusive story in @WIRED.

[Radical Numerics @RadicalNumerics]

@WIRED 11/ Our next campaigns expand Omnii-1’s capabilities to taxonomic tree traversal and functional prediction: locating unfamiliar sequences within the viral tree of life and predicting what they could do, from short reads toward whole genomes.

[Eric Nguyen @exnx]

Designing phages with AI is one challenge. Getting it to a patient is another.

Behind every treatment is a network of scientists, physicians, manufacturers, and partners working together.

Our new blog explores the journey from AI-generated DNA to patient care, and what it will take to scale it.

Blog: https://t.co/B3mCHgdUHe

[Eric Nguyen @exnx]

A few years ago, we were just a group of researchers chasing unconventional bets in AI.

That journey took us from new architectures like Hyena, to Evo and creating the field of generative genomics, to a much bigger realization.

We need AI that can understand biology as a whole. We started @RadicalNumerics to pursue that vision.

Here’s a conversation with my cofounders on how we got here, why we took the leap, and what we believe comes next.

[Radical Numerics @RadicalNumerics]

8/ With Omnii Diagnostics, we showed that pretraining across biological modalities transfers to a new biomarker, extending the range of Omnii applications.
Further studies are needed to confirm detection gains in larger screening populations.

We continue to work with our partners to build the next generation of diagnostics with Omnii.

https://t.co/e273SQH7KF

[Radical Numerics @RadicalNumerics]

1/ With Omnii Diagnostics, we asked whether a multimodal genome language model could help find cancers whose signals in blood are too faint for current classifiers.

We adapted Omnii-1 to a biomarker never seen in pretraining. In a test cohort, Omnii's features improved early-stage cancer detection

[Radical Numerics @RadicalNumerics]

7/ One question is whether Omnii-1 leaned on risk proxies like age or sex.

If so, gains would likely cluster in some groups. Instead, at a fixed threshold, its features raised early-stage sensitivity for women, men, and every age group tested.