Primate genomes analyzed by AI for clinical relevance of individual gene variants.
New genome data from a large number of different monkey species generated by an international research team are providing new insights into the genetic causes of human diseases. With development of a deep-learning-based algorithm, they could serve as a basis for personalized medical treatment concepts in humans in the future. Martin Kuhlwilm, a scientist at the Department of Evolutionary Anthropology at the University of Vienna, has been involved in this and other studies on the genome of various primates, the collected results of which have recently been published in the latest issue of the renowned journal Science. The results obtained are promising and provide further insights into the evolution of primates, into some (supposed) unique features of humans as well as into genetic causes of human diseases.
Already the talk of the town, but still very much up for development: personalized medicine based on the genetics of the individual, which should offer even more meaningful solutions and therapy concepts for people in the future. The prerequisite for answering this question with increasing precision is to identify, among the hundreds of thousands of mutations in the human genome, those that actually cause disease. Because progress in decoding the individual genome is slow and human data sets are only available over a limited period in the history of the earth, it was obvious to make use of information from the genome of closely related primate species: After all, the genetic diversity found in the 520 known non-human primate species is the evolutionary result of "nature’s experiments" on genetic variation that are constantly underway. These have been taking place continuously for millions of years - and thus provide a valuable source of data.
Genome from the catalog
First, the scientists sequenced recent and fossil primate genomes to create the most comprehensive catalog of nonhuman ape genome data to date, incorporating the results of 809 individuals from 233 species studied. The publication of this unique dataset thus includes genome data from nearly half of all primate species living on Earth. The data initially allowed the research team to compare species in phylogenetic analyses to improve understanding of primate evolutionary history.
But not only that - because based on the assumption that gene variations in one primate species have similar effects as the same variant in another species, population screenings were carried out in closely related species and common gene variants were systematically catalogued. Martin Kuhlwilm explains, "The new genome catalog has halved the number of genomic variations thought to be unique to humans. This reduction makes it easier to identify those mutations that we don’t share with primates and thus could be the basis for traits that make us human." "However, because we do not carry many of the recent genetic changes alone, but share them with other species, what actually makes us human appears to be rarer than expected," Kuhlwilm continues. "This shows that studying our living relatives helps us better understand our own species."
AI for disease-causing mutations.
By comparing the genomes of those 233 species of non-human primates with the human genome, more than 4.3 million "mis-mutations" in the genes could be identified, which subsequently change the composition of amino acids in the proteins and can be potential causes of human diseases. The actual identification of disease-causing mutations was achieved through the development of the deep-learning-based algorithm "PrimateAI-3D" and conversely allowed to exclude gene variants that occur together in humans and monkeys and are not pathogenic. Thus, 6% of the possible mismatches occurred frequently in primates and are therefore considered "potentially benign" because their presence is tolerated in these animals without signs of disease.
Considering that today more than 60% of the primate species on earth are already threatened with human-caused extinction in the next ten years, every effort should be made to protect these irreplaceable species. Because, says study author Lukas Kuderna, "We need to decide now whether we want to act to preserve these precious species. Each one is valuable in its own right - and our mirror for understanding our genomes and thus ourselves."
Original publications:
Kuderna, L.F.K et al, "A global catalog of whole-genome diversity from 233 primate species". Science. May, 2023.
abn7829
Sorensen, E, et al, "Genome-wide coancestry reveals details of ancient and recent male-driven reticulation in baboons". Science. May 2023.
abn8153
Gao, H, et al, "The landscape of tolerated genetic variation in humans and primates." Science. May 2023. abn8197
