DeepMind’s AI checks if DNA Changes are bad
This AI tool could become a valuable resource for geneticists and clinicians, offering more accurate insights into disease-causing mutations and potentially leading to improved treatments for various disorders.
Facebook
Twitter
LinkedIn
Pinterest
Pocket
WhatsApp
Scientists at Google DeepMind created a computer program that helps identify if genetic changes can make people sick. This program focuses on ‘missense mutations,’ where a single letter in our DNA is spelled wrong. Sometimes these mutations are fine, but other times they can lead to diseases like cystic fibrosis, cancer, and brain problems.
They used the program, called AlphaMissense, to look at all the different single-letter changes in our DNA that might affect how our proteins work. When the program was set to be very accurate, it said that 57% of these changes were probably okay, 32% were likely to be bad, and it wasn’t sure about the rest.
The scientists have shared their findings and created a free online tool for geneticists and doctors. This tool helps them understand which genetic changes might be causing diseases or disorders.
Usually, a person has around 9,000 genetic changes in their DNA. Out of over 4 million such changes in humans, only 2% have been classified as either harmless or harmful. Doctors already use computer programs to guess which changes might cause disease, but these guesses are not always right. They can only give some clues for making a diagnosis.
In their report in the journal Science, Dr. Jun Cheng and his team explain how AlphaMissense works better than the current programs that predict the effects of genetic changes. This should help experts figure out more quickly which changes are causing diseases. The program might also discover new changes linked to specific disorders and suggest better treatments.
AlphaMissense is based on DeepMind’s AlphaFold program, which can guess the 3D shape of proteins from their chemical building blocks.
To learn which genetic changes are common and probably harmless, AlphaMissense was given data on DNA from humans and similar animals. Same time, it learned about proteins by studying millions of protein sequences & figuring out what a healthy protein looks like.
When the AI is given a mutation, it gives it a score to show how likely it is to be a problematic change in the DNA. It can’t tell us exactly how this change causes issues.
Dr. Jun Cheng, one of the researchers, compared it to changing a word in a sentence. If you replace a word in an English sentence, someone who knows English can quickly tell if the change will affect the meaning of the sentence or not.
Professor Joe Marsh, a scientist at Edinburgh University who wasn’t part of this research, thinks that AlphaMissense has a lot of promise.
“We have this issue with computational predictors where everybody says their new method is the best,” he said. “You can’t really trust people, but [the DeepMind researchers] do seem to have done a pretty good job.”
If doctors and experts believe that AlphaMissense is trustworthy, its predictions could become more important for diagnosing diseases in the future,” he explained.
Professor Ben Lehner, who works on human genetics at the Wellcome Sanger Institute, mentioned that AlphaMissense’s predictions should be checked by other scientists. However, it seems to be good at figuring out which DNA changes lead to diseases and which ones don’t.
“One concern about the DeepMind model is that it is extremely complicated,” Lehrer said. “A model like this may turn out to be more complicated than the biology it is trying to predict. It’s humbling to realise that we may never be able to understand how these models actually work. Is this a problem? It may not be for some applications, but will doctors be comfortable making decisions about patients that they don’t understand and can’t explain?
“The DeepMind model does a good job of predicting what is broken,” he added. “Knowing what is broken is a good first step. But you also need to know how something is broken if you want to fix it. Many of us are very busy generating the massive data needed to train the next generation of AI models that will tell us not only which changes in DNA are bad but also exactly what the problem is and how we might go about fixing things.”
Facebook
Twitter
LinkedIn
Pinterest
Pocket
WhatsApp
Never miss any important news. Subscribe to our newsletter.
Related News
Tech to Reverse Climate Change & Revive Extinct Species
December 23, 2024
AI Unlocks the Brain’s Intelligence Pathways
December 23, 2024
XPENG Unveils Iron Robot with 60 Human-like Joints
November 11, 2024
Can AI Outsmart Humanity?
November 5, 2024
11 ChatGPT Prompts to Boost Your Personal Brand
October 30, 2024
Keir Starmer Hints at Possible Tax Hikes on Asset Income
October 25, 2024
Navigating the Future of AI: Insights from Eric Schmidt
October 24, 2024
AI Narrative Attacks: 3 Defense Strategies for Business Leaders
October 24, 2024
Your Health, Your Wealth
October 15, 2024
The Kashmiri Illustrator Transforming Art into Impact
October 15, 2024
9 Tips for Small Business Growth Success
October 15, 2024