In April 2003, scientists sequenced the first-ever human genome. That was a remarkable breakthrough at the time; finally being able to make sense of all the data that makes human life possible. Nearly 15 years later, it’s still not a simple task to complete, but at least now we have the likes of machine learning (ML) and artificial intelligence (AI) to help.
Google’s just launched a new AI tool called Deep Variant that makes sense of a person’s genome through sequencing data. It automatically detects mutations and gives scientists a much clearer picture as to the ins and out of the genome. While there are some tools around already that interpret these readouts, none are 100% accurate, and most rely on less advanced AI techniques.
“One of the challenges is in difficult parts of the genome, where each of the [tools] has strengths and weaknesses,” said Brad Chapman, a research scientist at Harvard’s School of Public Health. “These difficult regions are increasingly important for clinical sequencing, and it’s important to have multiple methods.”
DeepVarient was developed using millions of sequenced genomes from the Genome in a Bottle (GIAB) project by researchers from the Google Brain Team and others from Verily, a life sciences Alphabet subsidiary.
“The success of DeepVarient is important because it 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. DeepVarient will be accessible via the Google Cloud Platform in an attempt to lure more users to the platform.
Deep Genomics is a company using AI to identify genetic causes of diseases and create potential treatments from that. It uses deep learning to look for patterns in medical data to develop potentially life-changing drugs. Wuxi Nextcode is another big name in the industry and one that has invested heavily in machine learning and AI in the past few years.
There are many instances where AI has already proved to be invaluable in the medical field and it will continue to do so for a long time. And genomic data is even more invaluable. “For the first time in history, our ability to measure our biology, and even to act on it, has far surpassed our ability to understand it,” said Frey. “The only technology we have for interpreting and acting on these vast amounts of data is AI.”