- Author, Pallab Ghosh
- Role, BBC
DeepMind, a subsidiary of Google, uses artificial intelligence technology to identify changes in human DNA that cause diseases.
Scientists believe they have succeeded in identifying 89 percent of all major genetic changes.
This development is expected to enhance disease diagnosis and help in the search for better treatments.
Leading scientist Ewan Burney, deputy director-general of the European Molecular Biology Laboratory, told . that this work was a “major advance”.
He added: “This will help scientists conducting practical experiments determine research priorities in order to monitor areas that could cause disease.”
This technology works by checking the arrangement of components in DNA strands in humans.
All living organisms contain DNA, which is composed of four groups of chemicals called adenine (A), cytosine (C), guanine (G), and thymine (T). In the case of humans, as the fetus develops, the order of these letters is read to produce proteins, which are the building blocks of cells and tissues that make up different parts of the body.
However, if there is an incorrect arrangement of these letters, perhaps due to a genetic defect, the body’s cells and tissues do not form correctly, which leads to disease.
Last year, DeepMind’s artificial intelligence technology succeeded in determining the shape of almost all proteins found in the human body.
The new system, called alphamycin, can tell whether letters in DNA will produce the correct form or not, and if not, then see it as a potential disease-causing agent.
Geneticists currently have somewhat limited knowledge about the areas of DNA in humans that can lead to disease, and have only been able to classify 0.1 percent of letter changes or mutations as benign or disease-causing.
Pushmeet Kohli, of Google’s DeepMind, said the new model raised this percentage to 89 percent.
He added that scientists currently have to identify the areas that are likely to cause diseases across the billions of chemical building blocks that make up DNA, but the situation has changed now.
“Scientists can now focus their efforts on new areas that they were not aware of and that we have highlighted as potential pathogens,” Kohli said.
The new research, published in the journal Science, was tested by Genomics England, which cooperates with the British Health Insurance Authority.
Ellen Thomas, deputy chief medical officer at Genomics England, said the health service would be among the first organizations to benefit from the new development.
She added: “The new tool provides a new perspective on the data. It will help scientists who conduct practical experiments in understanding genetic data, in a way that benefits patients and medical teams.”
Bernie said he expects artificial intelligence to play a prominent role in molecular biology and life sciences.
He added: “I don’t know where things will go, but it changes almost everything we do at the moment.”