Jun Cheng of Google DeepMind said, “We should emphasize that the predictors were never actually trained on or intended to be used solely for clinical diagnosis.”
“However, we think our predictions could be helpful in increasing rare disease diagnosis rates, and potentially even help us find new disease-causing genes,” Cheng said.
Indirectly, this could lead to the development of new treatments, the researchers said.
The tool was trained on DNA from humans and closely-related primates, helping to identify which genetic mutations are widespread.
Cheng said the training allowed the tool to “input millions of protein sequences and learn what a regular protein sequence looks like.”
It can then identify the mutation and its potential for harm.
Cheng compared the process to learning a language.
“If we substitute a word from an English sentence, a person familiar with English can immediately see whether the word substitution will change the meaning of the sentence.”