New AI model for DNA learns from evolution to unlock secrets of the human genome

TL;DR

Summary:
- Researchers at UC Berkeley have developed a deep learning model called "DNABERT-2," which utilizes transformer-based architecture to interpret the complex "language" of the human genome.
- The model functions similarly to large language models like GPT, but instead of text, it analyzes DNA sequences to identify functional elements, predict the impact of genetic mutations, and provide insights into evolutionary biology.

Like summarized versions? Support us on Patreon!