back Back

AI-driven protein engineering

Nov. 27, 2024.
2 mins. read. 10 Interactions

A new AI system called EVOLVEpro can take existing proteins and make them more stable, accurate, and efficient.

About the Writer

Giulio Prisco

92.57456 MPXR

Giulio Prisco is Senior Editor at Mindplex. He is a science and technology writer mainly interested in fundamental science and space, cybernetics and AI, IT, VR, bio/nano, crypto technologies.

Nature has always been adept at creating proteins, but scientists have often found ways to enhance them. Now, Artificial Intelligence (AI) is stepping in with the potential to make proteins even more effective.

AI can be used to design proteins that could help in medicine, like creating antibodies that target diseases better or vaccines that work more efficiently against viruses. But the use of designer proteins isn’t just limited to healthcare; they could also be used to grow crops that are more nutritious or help the environment by absorbing carbon dioxide.

A new tool called EVOLVEpro, developed by researchers from Mass General Brigham and Beth Israel Deaconess Medical Center, is at the forefront of this AI-driven protein engineering. According to a study published in Science, EVOLVEpro can take existing proteins and make them more stable, accurate, and efficient.

For example, with EVOLVEpro, scientists can enhance how well a protein binds to its target, which is crucial for treatments like cancer or autoimmune diseases.

Large language models (LLMs) in AI are changing the game in protein engineering. These models work like AI systems that predict and generate human language, but for proteins, they predict how to tweak the protein “language” to get better results. EVOLVEpro uses this technology by building upon existing models to offer more refined suggestions for protein improvement.

The researchers have shown that EVOLVEpro can dramatically enhance proteins.

Protein engineering can transcend nature’s capabilities

The researchers believe that with tools like EVOLVEpro, we’re entering a new phase where protein design can transcend nature’s capabilities, potentially solving modern problems like climate change and food security by designing proteins tailored for these specific challenges.

In related news, other scientists have developed a method for designing large new proteins and producing them with the desired properties in the laboratory. Their approach involves a new way of using AlphaFold2, for which the Nobel Prize in Chemistry was awarded in 2024.

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter

Comment on this article

0 Comments

0 thoughts on “AI-driven protein engineering

7

Like

Dislike

Share

Comments
Reactions
💯 💘 😍 🎉 👏
🟨 😴 😡 🤮 💩

Here is where you pick your favorite article of the month. An article that collected the highest number of picks is dubbed "People's Choice". Our editors have their pick, and so do you. Read some of our other articles before you decide and click this button; you can only select one article every month.

People's Choice
Bookmarks