Evo and regLM are genome-focused AI models designed to generate and interpret sequences of mobile DNA and regulatory elements, responding efficiently to short prompts. These models embody a significant advancement in bioinformatics, aiming to decode the non-coding regions of the genome, which are crucial for gene expression regulation. Unlike predictable protein folding, the functionality of non-coding DNA sequences can be complex, overlapping, and context-dependent, offering a grand challenge in biological research akin to that presented by protein structure prediction.
AI models like Evo and regLM can generate new sequences of mobile DNA and predict regulatory activity in human cell lines based on prompts.
These tools aim to decode and interpret non-coding portions of the genome, which pose a significant challenge in understanding gene expression.
Collection
[
|
...
]