Advertisement

AI-powered genomic analysis reveals unknown human ancestor

"Deep learning allows us to observe what makes the ancestral puzzle fit together," said researcher Oscar Lao.

By Brooks Hays
In the diagram, Xe represents the unknown ancestral species, the product of interbreeding between two ancient human populations. Photo by Center for Genomic Regulation
In the diagram, Xe represents the unknown ancestral species, the product of interbreeding between two ancient human populations. Photo by Center for Genomic Regulation

Jan. 17 (UPI) -- Using a combination of deep learning algorithms and advanced statistical techniques, researchers identified an unknown human ancestor hiding in the modern human genome.

According to the new genomic analysis, a hybrid species produced by Neanderthals and Denisovans bred with Out of Africa modern humans in Asia some 40,000 years ago.

Advertisement

The discovery, detailed this week in the journal Nature Communications, marks the first time scientists have used deep learning algorithms to analyze human evolution.

The statistical analysis suggests hybrid hominids may have regularly interbred with modern humans.

"About 80,000 years ago, the so-called Out of Africa occurred, when part of the human population, which already consisted of modern humans, abandoned the African continent and migrated to other continents, giving rise to all the current populations," Jaume Bertranpetit, principal investigator at the Institute of Evolutionary Biology, said in a news release.

"We know that from that time onwards, modern humans cross bred with Neanderthals in all the continents, except Africa, and with the Denisovans in Oceania and probably in Southeast Asia, although the evidence of cross-breeding with a third extinct species had not been confirmed with any certainty," Bertranpetit said.

Advertisement

Scientists have previously theorized that a third species accounts for the origins of genomic fragments belonging to modern humans.

Deep learning algorithms mimic the mammalian nervous system, using a combination of artificial neurons to analyze data and detect patterns important to the performance of a specific task.

"We have used this property to get the algorithm to learn to predict human demographics using genomes obtained through hundreds of thousands of simulations," said Oscar Lao, principal investigator at the National Genomic Analysis Center in Barcelona. "Whenever we run a simulation we are traveling along a possible path in the history of humankind. Of all simulations, deep learning allows us to observe what makes the ancestral puzzle fit together."

Last summer, paleontologists recovered the remains of a hybrid hominid -- the offspring of a Neanderthal mother and a Denisovan father. The latest findings seem to confirm the fossil's identification.

"Our theory coincides with the hybrid specimen discovered recently in Denisova, although as yet we cannot rule out other possibilities," said Mayukh Mondal, researcher at the University of Tartu.

Latest Headlines