Scientists now have access to Evo 2, a groundbreaking AI foundation model trained on 9 trillion nucleotides—the largest publicly available AI model for genomic data. Built using NVIDIA DGX Cloud on AWS, Evo 2 is a collaboration between the Arc Institute and Stanford University, offering unparalleled insights into DNA, RNA, and proteins across species.
The model is now accessible through NVIDIA BioNeMo, with support for NVIDIA NIM microservices, allowing seamless AI deployment for researchers worldwide.
🧬 Key Applications of Evo 2:
✔ Predicting protein structure and function based on genetic sequences
✔ Identifying novel molecules for medicine, agriculture, and materials science
✔ Analyzing gene mutations to study disease mechanisms
🚀 “Evo 2 represents a major milestone for generative genomics,” said Patrick Hsu, Arc Institute cofounder and professor at UC Berkeley. “By advancing our understanding of the fundamental building blocks of life, we can unlock solutions in healthcare and environmental science that were previously unimaginable.”
Accelerating Scientific Discovery With AI
The Arc Institute, founded in 2021 with $650 million in funding, provides long-term support for pioneering scientific research. Its investigators work in disease areas like cancer, neurodegeneration, and immune dysfunction, leveraging NVIDIA’s cutting-edge AI technologies to push the boundaries of biomolecular science.
Powering Evo 2 With NVIDIA DGX Cloud
🔹 2,000 NVIDIA H100 GPUs enabled faster model training and scaling
🔹 DGX Cloud on AWS provided on-demand access to massive compute clusters
🔹 NVIDIA BioNeMo optimized AI workflows with NIM microservices and Blueprints
🌍 Impact: NVIDIA’s AI expertise helped researchers analyze larger datasets, develop new algorithms, and achieve faster breakthroughs in biomolecular research.
Transformative Applications Across Biomolecular Sciences
1. Healthcare & Drug Discovery
🔬 Understanding Disease-Causing Mutations – Evo 2 can predict which gene variants are linked to diseases like cancer.
💊 Accelerating Drug Discovery – AI can design targeted molecules to treat genetic disorders.
✔ Case Study: Evo 2 analyzed BRCA1 (breast cancer gene) mutations with 90% accuracy, predicting which previously unrecognized mutations affect gene function.
2. Agriculture & Global Food Security
🌱 Optimizing Crop Genetics – AI can help develop climate-resilient, nutrient-dense crops.
🌾 Tackling Global Food Shortages – Insights from Evo 2 could enhance agricultural biotechnology.
3. Sustainable Materials & Environmental Science
🛢️ Engineering Enzymes to Break Down Plastics & Oil – AI models can design proteins for environmental cleanup.
⚡ Developing Biofuels – Evo 2 can assist in creating sustainable energy solutions.
🔍 “Deploying Evo 2 is like launching a powerful new telescope into deep space,” said Dave Burke, CTO at Arc. “We know there’s immense opportunity for discovery, but we have yet to uncover what’s possible.”
Unlocking the Future of Biomolecular AI
Evo 2 is available now through NVIDIA BioNeMo, allowing researchers to:
✅ Fine-tune Evo 2 on proprietary datasets using the BioNeMo Framework
✅ Deploy AI models easily with NVIDIA NIM microservices
✅ Advance research in genomics, medicine, and sustainability
🔗 Get started with Evo 2 today on NVIDIA BioNeMo! 🚀
