Google on Monday released DeepVariant, an artificial intelligence (AI) tool that uses gene sequencing data to build a more accurate genomic model by automatically spotting small insertion and deletion mutations and single-base-pair mutations.
The Google Brain team compiled millions of high-throughput reads and fully sequenced genomes from the Genome in a Bottle project, feeding the data to a deep-learning system and modifying the parameters of the model until it learned to interpret sequenced data with very high accuracy.
"DeepVariant...demonstrates that in genomics, deep learning can be used to automatically train systems that perform better than complicated hand-engineered systems," says Deep Genomics CEO Brendan Frey.
Frey predicts AI will ultimately transcend its ability to help sequence genomic data. "The gap that is currently blocking medicine right now is in our inability to accurately map genetic variants to disease mechanisms and to use that knowledge to rapidly identify life-saving therapies," he says.
From Technology Review
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA