Researchers have developed a ground-breaking machine-learning method that employs computer systems to seek patterns and trends in data sets to determine whether medications like dextromethorphan could be repurposed to aid people in quitting smoking. These findings were published in Nature Genetics.

Previous studies identified that certain persons with particular genes are more prone to develop a tobacco addiction. Researchers in the current study used genetic data from more than 1.3 million people from an extensive multi-institution study involving machine learning to evaluate these large data sets about a person′s genetics and self-reported smoking practices.

More than 400 genes were identified to be linked to smoking behaviours by the researchers. They had to figure out why certain of those genes were associated with smoking behaviours. As a person can have thousands of genes, so the researchers had to analyze the reason behind only some genes exhibiting smoking behaviours. There were clear associations between genes that hold instructions for producing nicotine receptors or are involved in signalling for the hormone dopamine, which helps people feel calm and content. The research team had to identify each gene′s function in biological pathways for the remaining genes, and then use that knowledge to establish which drugs are already authorized for altering those existing pathways.

The study showed that since people with European ancestry made the majority of the study′s genetic data, the machine learning model had to be modified to analyze that data as well as a smaller data set of about 150,000 persons with Asian, African, or American ancestry.

The study screened out east eight drugs, including dextromethorphan, and galantamine, as having the potential to be repurposed for smoking cessation.

Experts concluded that the findings would enhance the precision with which machine learning models can identify people at risk for drug abuse and identify potential biological pathways that can be focused on for beneficial therapies.

(Source: https://theprint.in/health/machine-learning-identifies-drugs-to-help-smokers-quit/1344094/)