Deep Learning Finds a New Human Ancestor

Researchers combined statistical methods with deep learning algorithms to find an unknown human ancestor.

Investigators from the Institute of Evolutionary Biology (IBE), the Institute of Genomics at the University of Tartu, and the Center for Genomic Regulation (CRG) found the footprint of a new human ancestor. The computational analysis of the DNA revealed that the extinct species was a hybrid of Denisovans and Neanderthals. This evidence will help scientists to understand the recent finding of a hybrid (the offspring of a Denisovan father and Neanderthal mother) in the caves of Denisova. Prior to this research, it was believed that this discovery was an isolated case. However, this latest analysis is clear proof of the fact that it was part of a more general introgression process.

One of the ways of distinguishing two cross-breeding species is to have a look at the fertility of their descendants because they mostly do not produce fertile descendants. This is already a difficult concept to understand but it becomes even harder when extinct species are involved. The current study blurred the lines of these limits even further as it suggested that hominids (Denisovans and Neanderthals) coexisted with humans around 40,000 years ago in Eurasia. Jaume Bertranpetit, the Principal Investigator of the latest research at the Institute of Evolutionary Biology (IBE) of CRG, elaborated the findings in the following words:

“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. 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 South-East Asia, although the evidence of cross-breeding with a third extinct species had not been confirmed with any certainty.”

Scientists were looking for the existence of the third human ancestor because it could explain the origin of some fragments of the current genome of humans. Deep learning allowed the researchers to make the transition from DNA to the demographics of ancestral populations. Having said that, the complexity of these demographic models was too much even for the most advanced statistical tools, which created some problems for the investigators. Oscar Lao, an Expert of Simulations and the Principal Investigator at the CNAG-CRG, talked about the importance of deep learning and its usage in their investigation. He said,

“[Deep Learning] is an algorithm that imitates the way in which the nervous system of mammals works, with different artificial neurons that specialize and learn to detect, in data, patterns that are important for performing a given 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. 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.”

Deep learning belongs to a family of machine learning methods based on learning data representations. Some common architectures of deep learning include deep neural networks, recurrent neural networks, and deep belief networks. The models used for deep learning are vaguely inspired by communication patterns and information processing in biological nervous systems. However, there are certain differences in structural and functional properties, which make them incompatible with neuroscience findings.

It is the first occasion when artificial intelligence has proved successful in finding a new human ancestor. Scientists believe that this breakthrough will pave the way for this technology to be used by other fields (Genomics, Biology, and Evolution) to answer many important questions about human history. The deep learning analysis of the DNA suggested that the extinct hominid is possibly a descendant of the Denisovan and Neanderthal populations. The fossil discovered recently from Denisova also seems to support the idea that a third species coexisted with modern human beings and gave birth to a hybrid. Mayukh Mondal, a Former Investigator at the IBE who is currently investigating at the University of Tartu, referred to that by saying,

“Our theory coincides with the hybrid specimen discovered recently in Denisova, although as yet we cannot rule out other possibilities.”