Scientists are currently exploring ways to make our food supply more climate-resilient. To do this, they want to cultivate plant "superpowers," like the ability to survive severe droughts or yield more fruit, by finding the specific variants responsible for those advantages.
But… there’s a catch — traditional software can't tell which DNA sequences actually alter the trait and which DNA sequences are just harmless lookalikes on the same chromosome.
To actually find the answers they’re looking for, scientists usually have to physically breed, grow, and test thousands of plants over multiple seasons to (VERY slowly) separate the traits. This physical trial-and-error can take years or even decades. Until now.
AI research lab Living Models paired @GoogleGemma 4 with BOTANIC-1, a specialized AI trained on plant DNA, to rapidly pinpoint the exact mutations that make crops thrive.

How is this happening? Gemma 4 acts as a sort of project manager: writing code, organizing data, and filtering out the irrelevant information that gets in the way of finding the real answer. It hands the genetic puzzles to BOTANIC-1, which has been trained on 320 plant species and understands millions of years of evolution so it can pinpoint which DNA changes are most likely to alter the plant.
In a test for a gene that could positively impact melon yields, older tools got stuck searching between thousands of mutations. Gemma and BOTANIC-1 found the exact match in under four minutes to correctly rank the target mutation at #1 out of 2,494 possibilities.

But this is about way more than melons. As climate shifts disrupt traditional farming, this AI pipeline could potentially give agricultural researchers the speed they need to engineer climate-resilient crops and secure the global food supply on the timeline at hand.
Download BOTANIC-1 →https://goo.gle/4y7qDak
Read the paper →https://goo.gle/4xWrKJB
Try the Gemma 4 + BOTANIC-1 demo →https://huggingface.co/spaces/living-models/gemma-botanic
And check out the blog to learn more: https://deepmind.google/models/gemma/gemmaverse/living-models/





