New AI Tool ESM Atlas Expands Protein Structure Database Significantly
New protein-folding AI vastly expands on Alphafold's efforts
Image: Scientificamerican
The Chan Zuckerberg Initiative's Biohub has launched the ESM Atlas, an AI-generated database containing over one billion predicted protein structures, surpassing AlphaFold's database. Utilizing the ESMFold2 model, this atlas aims to enhance biological discoveries by providing insights into both known and unknown proteins.
- 01The ESM Atlas includes 1.1 billion predicted protein structures and 6.8 billion protein sequences, significantly exceeding AlphaFold's offerings.
- 02ESMFold2 outperforms AlphaFold3 in predicting protein interactions, aiding in the design of new antibodies for cancer and immunological conditions.
- 03The atlas will be freely accessible, promoting collaboration and research across various scientific fields.
- 04Scientists are optimistic about the potential of ESMFold2, particularly its ability to analyze metagenomic sequences not covered by AlphaFold.
- 05The open-source nature of ESMFold2 allows for unrestricted commercial use, increasing its potential adoption in research.
Advertisement
In-Article Ad
The Chan Zuckerberg Initiative’s Biohub has unveiled the ESM Atlas, a groundbreaking AI tool that predicts over 1.1 billion protein structures and 6.8 billion protein sequences, far exceeding the previous AlphaFold database. Developed using the ESMFold2 model, which is said to surpass AlphaFold3, the atlas is designed to facilitate biological discoveries by illuminating both known and previously uncharacterized proteins. Alex Rives, head of Biohub science, highlights the atlas as a comprehensive resource for exploring protein biology. ESMFold2, based on a 'protein language' model trained on diverse metagenomic sequences, excels in predicting the structures of interacting proteins, including antibodies targeting cancer-related proteins. While the scientific community has expressed excitement about the atlas, some experts emphasize the need for further evaluation of its predictions. The fully open-source nature of ESMFold2 is expected to encourage widespread use and innovation in protein research, potentially leading to new insights in biology and medicine.
Advertisement
In-Article Ad
The ESM Atlas is expected to significantly enhance research capabilities in protein biology, potentially leading to breakthroughs in medical and environmental applications.
Advertisement
In-Article Ad
Reader Poll
How do you think the ESM Atlas will impact protein research?
Connecting to poll...
Read the original article
Visit the source for the complete story.