Introducing Axis: the first AI model that generates regulatory DNA elements and predicts their function. Gene therapies suffer from poor efficacy, toxicity & specificity. Models like Axis can help overcome such risks. Axis beats Google DeepMind's AlphaGenome at predicting

Read more about Axis here: https://t.co/bYk6CkLQXO

Interested in knowing more about how we beat Google DeepMind? We shared some insights on @tbpn

Read more about Axis in our blog here:

On TBPN live, our co-founder and CEO shares more on how we plan to use Axis to develop new age gene therapies👇

Going on TBPN at 1:45 pm PT to launch @origin_bio and our model, Axis. Axis is the first model that generates regulatory DNA sequences and predicts its function. Axis already outperforms DeepMind's AlphaGenome on various benchmarks!

We’re excited to take the model into the lab and open up public access soon! To get early access to the model, please fill out the form here: https://t.co/I3wIwYHxfU If you enjoyed this thread and want to learn more about Axis, read our blog post:

Natural genomes capture only a fraction of possible sequence designs permitted by underlying rules. Models like Axis explore this design space and can generate regulatory sequences that encode circuit-like logic, enabling cell type and state-specific activity.

Training the model to generate DNA at a high resolution resulted in overall improvements in its ability to predict the function of unseen sequences. This multifunctional design exposes Axis to different dynamics, unlocking better generalization capabilities.

Genes are turned “on” or “off” by TFs that bind to specific sites on the genome. In silico tests of the model’s generated sequences showed up to a 9x increase in the presence of these sites when prompted with the respective TFs.

Axis is promptable. Provide a regulatory element type (promoter or enhancer) with transcription factor (TF) proteins and it generates candidate sequences that enable the binding of those proteins.