An artificial intelligence (AI) model has simulated half a billion years of molecular evolution to design the blueprint for a previously unknown protein called esmGFP. This glowing protein, akin to those found in jellyfish and corals, could pave the way for groundbreaking medical advancements, researchers claim.

Tiffany Taylor, an evolutionary biologist at the University of Bath in the U.K., praised the potential of AI-driven protein engineering but expressed caution. "AI-driven protein engineering is intriguing, but I can't help feeling we might be overly confident in assuming we can outsmart the intricate processes honed by millions of years of natural selection," Taylor said.

She also noted that while the study simulates 500 million years of evolution, it focuses only on individual proteins, bypassing the broader stages of natural selection that create life.

The Role of Proteins in Life

Proteins are essential building blocks of life, performing diverse functions like building muscles and combating disease. The newly designed protein, esmGFP, represents a novel type of green fluorescent protein (GFP). In nature, GFPs are responsible for the luminescent glow of jellyfish and corals.

The computer-generated sequence for esmGFP is only 58% similar to its closest known fluorescent counterpart — a human-modified protein derived from bubble-tip sea anemones (Entacmaea quadricolor). The remaining 42% of the sequence, created by the AI model, would have required 96 genetic mutations and 500 million years to evolve naturally.

https://usaherald.com/tiktok-alternatives-threads-bluesky-and-more/

ESM3: The AI Behind esmGFP

Researchers at EvolutionaryScale unveiled esmGFP using an AI model called ESM3, introduced in a preprint study in 2024. 

Now peer-reviewed and published in Science, ESM3 operates differently from natural evolutionary processes. Instead of following evolution’s constraints, ESM3 solves protein-design problems by filling in gaps in incomplete protein sequences.

"We've found that ESM3 learns fundamental biology, and can generate functional proteins outside the space explored by evolution," said Alex Rives, co-founder and chief scientist of EvolutionaryScale, in an email to Live Science.

The development of ESM3 builds on earlier research conducted at Meta, the parent company of Facebook and Instagram, before EvolutionaryScale was founded in 2024. ESM3 is a generative AI model similar to OpenAI's GPT-4 but focuses on biology.

AI Develops Genetic Code

Proteins are made of chains of amino acids. And their sequences are encoded by genes. 

Each protein folds into a unique shape that determines its function. To train ESM3, researchers fed it data from 2.78 billion naturally occurring proteins. The AI was tasked with filling in missing parts of protein sequences, much like completing a sentence with missing words.

"The same way a person can fill in the blanks in the soliloquy 'to _ or not to _, that is the _,' we can train a language model to fill in the blanks in proteins," Rives explained.

Green fluorescent proteins like esmGFP are widely used in research labs, where their genetic code is often fused to other DNA sequences to track proteins and cellular processes. According to Rives, ESM3 could accelerate protein engineering for applications like drug development.

With esmGFP, ESM3 demonstrates the transformative potential of AI in protein engineering, creating novel proteins beyond the reach of natural evolution. While the innovation has sparked excitement, experts emphasize the importance of balancing optimism with recognition of nature’s complexity.