Artificial intelligence continues to revolutionize science, with recent advances now reaching the field of genetics.
Researchers have developed a powerful AI model capable of analyzing and designing genetic code at an unprecedented scale.
This technology has the potential to accelerate discoveries in medicine, agriculture, and synthetic biology, transforming our understanding of life itself.
AI Model Trained on 100,000 Species
As reported by HotNews.ro, the Arc Institute, in collaboration with NVIDIA and researchers from Stanford, UC Berkeley, and UC San Francisco, has developed Evo 2, the largest AI model for biological research.
The system has been trained on DNA data from over 100,000 species, allowing it to recognize patterns in genetic sequences that would take human scientists years to identify.
Evo 2 can accurately detect disease-causing mutations in human genes and even design new genomes similar to those of simple bacteria.
It can also analyze genetic sequences of up to one million nucleotides at once, offering insights into long-range interactions within a genome.
A Major Leap in Genetic Research
Building on its predecessor, Evo 1, which was limited to single-cell genomes, Evo 2 has been trained on a staggering 9.3 trillion nucleotides and 128,000 complete genomes.
This immense dataset provides researchers with a powerful tool for studying the genetic basis of diseases and engineering new biological systems.
The model was trained over several months using NVIDIA DGX Cloud AI on Amazon Web Services, utilizing more than 2,000 NVIDIA H100 GPUs.
To encourage further advancements, Arc Institute has made Evo 2’s code publicly available on GitHub, hoping it will be applied in other fields to accelerate scientific progress.
As AI continues to merge with genetics, the potential for groundbreaking discoveries grows. With tools like Evo 2, scientists may soon unlock new ways to treat genetic disorders, improve crop resilience, and even create synthetic life forms.
