Artificial active

UW Medicine researchers are ‘training’ artificial intelligence to create vaccines and drugs

Researchers at the University of Washington School of Medicine and Harvard University have developed artificial intelligence software to create proteins used in health sciences.

Their findings were published in the journal Science, explaining that using machine learning, scientists can “train” artificial intelligence to produce new solutions to certain prompts. These proteins could potentially be used to make vaccines, cancer treatments or even remove carbon pollution from the air, their research suggests.

“The proteins we find in nature are amazing molecules, but engineered proteins can do so much more,” said David Baker, professor of biochemistry at UW Medicine and lead author of the study. “In this work, we show that machine learning can be used to design proteins with a wide variety of functions.”

Scientists usually create proteins manually on a computer. These include antibodies and synthetic binding proteins to fight COVID-19, or enzymes for industrial manufacturing.

The problem is that a single protein molecule is already incredibly complex, with thousands of bound atoms making them difficult to study, even with specialized software.

This is where the AI ​​comes in.

A screenshot of DALL-E 2 demonstrating the ability of AI to generate original images from a prompt (DALL-E 2 // OpenAI)

Several recent projects like DALL-E use text prompts to generate images, which has inspired researchers to use the same concept in medicine.

“The idea is the same: neural networks can be trained to see patterns in data,” said lead author Joseph Watson, postdoctoral researcher at UW Medicine. “Once trained, you can give it a prompt and see if it can generate an elegant solution. Often the results are compelling, even beautiful.”

The team “trained” several neural networks with information from the Protein Database, a public database of hundreds of thousands of protein structures from across the animal kingdom.

The researchers say their first approach to generating proteins is called “hallucination,” which works like AI image-generating tools, which create new output based on a simple prompt. The second approach is called “inpainting”, which works like auto-complete features you might find on search engines or text messages.

“Most people can find new pictures of cats or write a paragraph from a prompt if asked, but with protein design, the human brain can’t do what computers can do now,” said lead author Jue Wang, a postdoctoral researcher at UW Medicine. “Humans just can’t imagine what the solution might look like, but we have machines in place that do.”

To generate the protein, the research team compares it to an AI-generated book.

“You start with a random assortment of words – total gibberish. Then you make a requirement like in the opening paragraph it has to be a dark, stormy night. Then the computer will change the words one at a time and wonder ‘Does this make more sense of my story?’ If so, it keeps the changes until a full story is written,” Wang said.

Rather than an opening paragraph, the researchers start with a string of amino acids instead. The software then transforms the sequence again and again until it encodes the desired function, a process that sometimes only takes a few seconds.

These sequences can then be fabricated and studied in the laboratory.

“With auto-completion, or ‘Protein Inpainting’, we start with the key features we want to see in a new protein and then let the software come up with the rest. These features can be known binding motifs or even sites enzyme actives,” Watson said.

According to the researchers, proteins made by hallucination and inpainting worked, including proteins that bind metals, others that bind to the cancer receptor PD-1, and even more that could vaccinate against respiratory syncytial virus (RSV). ).

Extensive testing is needed before these proteins are deployed for medical use.

RELATED: New UW study on the effective delivery of abortion drugs via telehealth services

Get news alerts in the FREE FOX 13 Seattle app. To download for Apple iOS or Android. And sign up for BREAKING NEWS emails delivered straight to your inbox.

“These are very powerful new approaches, but there’s still a lot of room for improvement,” said Baker, who received the 2021 Breakthrough Award in Life Sciences. “Designing high-activity enzymes, for example, is always very difficult. But each month, our methods continue to improve! Deep learning has transformed protein structure prediction over the past two years, we are now in the midst of a similar transformation of protein design. “