Researchers have used advanced AI and large genomic datasets to figure out how humans have adapted to recent diseases. It turned out that the gene mutation that causes hereditary Mediterranean fever is still prevalent as a result of ongoing natural selection.

Natural selection is the process by which beneficial gene mutations persist from generation to generation until they become dominant in our genomes – the catalog of all of our genes. One thing that can stimulate natural selection is pathogen protection.

However, if a population of people moves from one environment to another or changes their lifestyle, gene mutations that protect against one pathogen can make people susceptible to new diseases.

Familial Mediterranean Fever (FMF) is one example of this new disease.

FMF is an autosomal recessive disorder characterized by recurrent attacks of fever and peritonitis, sometimes pleurisy, skin lesions, arthritis, and rarely pericarditis. Renal amyloidosis may develop, sometimes leading to renal failure. People with a genetic origin from the Mediterranean basin are more likely to suffer from this disease than other ethnic groups. The diagnosis is mostly clinical, although genetic testing is available.

This inherited autoimmune disease has emerged over the past 20,000 years. It is common in southern Europe, the Middle East and North Africa, where about 50% of people in the region today carry a gene mutation that makes them more susceptible to the disease.

This predominance of a seemingly harmful gene mutation may be the result of two different types of natural selection.

One of the options is “incomplete stripping”, when the mutation of the susceptibility gene is in the process of being removed from the population, but has not yet been completely eradicated. In this case, natural selection continues.
Another option is “balancing selection,” in which some potentially deleterious gene mutations for one condition persist in the population because they provide some protection against another disease. In this case, the FMF susceptibility gene was associated with protection against the plague-causing bacteria Yersinia pestis.
To determine which version of natural selection is involved in FMF, the researchers turned to advanced AI that is especially good for pattern detection or image recognition. They trained the algorithm on datasets with known values ​​to test its ability to spot patterns.

The scientists then ran their algorithm in the database for the 1,000 Genomes Project, which contains genomic data for 2,504 people from 26 populations, including those in the Mediterranean. They found that mutations in the FMF gene are still prevalent as a result of ongoing selection; they have not yet reached equilibrium, and natural selection is still at work.

“This is the first tool to test the difference between different types of natural selection. It detects signals in the genome that were previously unavailable, explains lead researcher Dr. Matteo Fumagalli of the Department of Life Sciences at Imperial College London. “Now that we have proven that AI can be used to search genomes for fine selection schemes, it will come in handy for further studying how humans adapt to both old diseases like the plague and relatively new ones like FMF and COVID-19.”

One area of ​​the disease that the team is currently investigating is the human relationship with coronaviruses. People have lived with coronaviruses for at least 50,000 years, and the greater susceptibility of some people to the more serious COVID-19 could signal a different balancing selection mechanism.