"Google" Launches AI Tool to Understand Disease-Causing Genetic Mutations

SadaNews - Researchers at "Google DeepMind" revealed their latest tools in the field of artificial intelligence, confirming that it will help scientists identify genetic factors that cause diseases, which could ultimately pave the way for the development of new treatments, according to what was reported by The Guardian.

The new tool, named "AlphaGenome", predicts how genetic mutations affect the regulation of genes, including when they are activated, the types of cells in which they are active within the body, as well as whether levels of biological control are high or low.

Most common genetic diseases, such as heart diseases, autoimmune disorders, and mental health issues, are linked to mutations that affect gene regulation, as is the case with many types of cancer. However, identifying the genetic defect responsible for these conditions remains a significant scientific challenge.

In a press conference about this work, Natasha Latysheva, a researcher at "DeepMind", said: "We see AlphaGenome as a tool for understanding the functions of various elements within the genome, and we hope that it will accelerate our basic understanding of what is known as the code of life."

The human genome consists of about 3 billion pairs of genetic letters - guanine, thymine, cytosine, and adenine - which form the DNA code. Only about 2% of the genome determines how proteins are produced, the building blocks of life, while the remaining part coordinates gene activity by carrying precise instructions that specify where, when, and to what extent each gene is activated.

The researchers trained the "AlphaGenome" program on public databases of human and mouse genetics, which enabled it to understand the connections between mutations in specific tissues and their effects on gene regulation. This artificial intelligence can analyze up to one million DNA code letters at a time and predict how mutations will affect various biological processes.

The "DeepMind" team believes that this tool will help scientists identify which part of the genetic code is most crucial for the development of specific tissues, such as neurons and liver cells, as well as identify mutations that have the most significant impact on the emergence of cancer and other diseases. It could also form the basis for developing new gene therapies by enabling researchers to design innovative DNA sequences - for example, to activate a specific gene in nerve cells without affecting muscle cells.

In this context, Karl de Boer, a researcher at the University of British Columbia in Canada who was not involved in this work, said: "AlphaGenome can determine whether mutations affect genome regulation, which genes are affected, how this occurs, and in which types of cells. Thus, medications targeting these effects can be developed."

He added: "The ultimate goal is to have models that are accurate enough that we don’t need to conduct further experiments to confirm their predictions. Although AlphaGenome represents a significant innovation, reaching this goal requires ongoing efforts from the scientific community."

Some scientists have already begun using the "AlphaGenome" tool. Mark Mansour, clinical professor of hematology and oncology for children at the University of California, Los Angeles, stated that it represents a "qualitative leap" in his research regarding the discovery of genetic triggers for cancer.

For his part, Gareth Hawks, a statistical geneticist at the University of Exeter, said: "The non-coding genome constitutes 98% of the human genome, which consists of 3 billion base pairs. We understand the 2% that is coding relatively well, but having a tool like AlphaGenome that can predict the functions of the remaining 2.94 billion base pairs is a tremendous step forward for us."