![]() If clinical experts decided that AlphaMissense was reliable, its predictions may carry more weight in future disease diagnosis, he said. We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. For more information see our Privacy Policy. Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. “You can’t really trust people, but do seem to have done a pretty good job.” skip past newsletter promotion “We have this issue with computational predictors where everybody says their new method is the best,” he said. ![]() Prof Joe Marsh, a computational biologist at Edinburgh University who was not involved in the work, said AlphaMissense had “great potential”. ![]() “If we substitute a word in an English sentence, a person familiar with English can immediately see whether the word substitution will change the meaning of the sentence or not.” “This is very similar to human language,” Cheng said. When the trained AI is fed a mutation, it generates a score to reflect how risky the genetic change appears to be, though it cannot say how the mutation causes any problems. At the same time, the program familiarised itself with the “language” of proteins by studying millions of protein sequences and learning what a “healthy” protein looks like. The AI is an adaptation of DeepMind’s AlphaFold program, which predicts the 3D structure of human proteins from their chemical makeup.ĪlphaMissense was fed data on DNA from humans and closely related primates to learn which missense mutations are common, and therefore probably benign, and which are rare and potentially harmful. The program may also flag mutations that have not previously been linked to specific disorders and guide doctors to better treatments. Writing in Science, Dr Jun Cheng and others describe how AlphaMissense performs better than current “variant effect predictor” programs and should help experts pinpoint more swiftly which mutations are driving diseases. Doctors already have computer programs to predict which mutations may drive disease but because the predictions are inaccurate, they can only provide supporting evidence for making a diagnosis. Of more than 4m seen in humans, only 2% have been classified as either benign or pathogenic. It was uncertain about the impact of the rest.īased on the findings, the scientists have released a free online catalogue of the predictions to help geneticists and clinicians who are either studying how mutations drive diseases or diagnosing patients who have rare disorders.Ī typical person has about 9,000 missense mutations throughout their genome. When they set the program’s precision to 90%, it predicted that 57% of missense mutations were probably harmless and 32% were probably harmful. The researchers used AlphaMissense to assess all 71m single-letter mutations that could affect human proteins.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |