Researchers from Southern Medical University developed an artificial intelligence (AI) application that identifies extensive tradeoffs in different mutations brought about by human evolution. Previous research showed that mutations occur as creatures evolve, and these mutations could lead to benefits or purely hitchhikers that remain in the genome because of their proximity to genes with benefits.
The group describes in their study, titled "Uncovering the Extensive Trade-off Between Adaptive Evolution and Disease Susceptibility," published in the journal Cell Reports, how they used data from existing genome-wide associations to teach the AI and what it shows when exposed to new data.
AI Used in Identifying Genomic Tradeoffs in Mutation Brought About Human Evolution
Role of Mutation in Evolution
According to the American Phytopathological Society (APS), the mutation results from a change in DNA and is the ultimate source of new alleles that create new genotypes. Small populations have fewer alleles because of the genetic drift and fewer mutations generated within a small population.
Mutation plays a critical role in evolution and is the source of all genetic variation. The first step of evolution is creating a new DNA sequence for a specific gene. Although recombination can also create a new DNA sequence for a particular gene, mutation acts as an evolutionary force by itself and could cause significant changes in the allele frequencies over time.
These mutations or gene variants result in genetic variations that alter gene activity or protein function, Medline Plus reports. If a trait is advantageous, it can help the individual to survive and reproduce and will likely be passed down to the next generation through natural selection.
As generations continue to reproduce, these traits become increasingly common in a population, making it different from its ancestors. But sometimes, these differences could also be considered new species.
On the other hand, some genetic variants do not influence evolution. They could occur during a person's lifetime but only in some of the body's cells and are not hereditary. Also, many genetic variants do not impact the gene's function and are neither helpful nor harmful.
READ ALSO: Blue-Eyed People Share One Ancestor That Developed the Mutation 6,000 to 10,000 Years Ago
AI System Capable of Finding Mutational Tradeoffs in Human Genome
Phys.org reported that the new study authors had developed an AI application that could determine which mutations in the human genome have benefits and which were only hitchhikers. They call this AI the DeepFavored and fed it with data from existing genome-wide association studies.
The team narrowed down the data into alleles related to diet and metabolic activities, as well as mutations that allowed for dealing with variations in climate or those that help humans adapt to the climate in different parts of the world.
Then, they ran through the AI application on three different human populations and identified the hitchhikers and those that increase the risk of developing certain diseases.
Researchers concluded in their study that the AI tool could find evidence for mutational tradeoffs in the human genome. They could use DeepFavored results to compare to two other algorithms, wherein the former outperformed the two.
RELATED ARTICLE: How AI Deep Learning Neural Networks Prevent Species Extinction and Protect Indigenous People
Check out more news and information on Genetics in Science Times.
![Earth's Quasi-Moon Kamo‘oalewa Could Originate From Lunar Surface Not Asteroid Belt [Study]](https://1721181113.rsc.cdn77.org/data/thumbs/full/53275/89/56/50/40/earths-quasi-moon-kamo-oalewa-could-originate-from-lunar-surface-not-asteroid-belt-study.png)
