The Nobel Prize in Physics recently won by John J. Hopfield and Geoffrey Hinton for their founding contributions to the field of machine learning, a branch of AI, with artificial neural networks, is joined by that in Chemistry awarded to David Baker, John Jumper and Demis Hassabis for his work on the AlphaFold project, a revolutionary artificial intelligence capable of predicting the three-dimensional structure that a protein will adopt. But AI is capable of much more.
Thanks to artificial intelligence, new antibiotics have been discovered, a supernova, thousands of new materialshas revealed archaeological secrets such as those of the Nazca lines and possible genetic and oncological treatments.
Its capabilities not only focus on being able to analyze a huge amount of data in record time, it is also capable of deducing relationships and establishing links, testing different strategies and evaluate the possibilities of treatments and drugs or model climate change in hundreds of different scenarios and with thousands of variables. But now he has gone one step further.
artificial intelligence has been used to reveal details of a diverse branch and fundamental of the life that lives right under our feet and in every corner of the planet.
Thanks to this technology, they have discovered 161,979 new species of RNA viruses. This was made possible through the use of a machine learning tool that scientists believe will greatly improve the mapping of life on Earth and could assist in the identification of many millions more viruses that have not yet been characterized. In fact, it is believed that there are more viruses on our planet than stars in the universe.
The results have been published in the journal Cell and the research has been carried out by an international team that has written “the largest virus species discovery study ever published”according to a statement.
“We have been offered a window into a part of life on Earth that would otherwise be hidden, revealing remarkable biodiversity – explains study leader Edwards Holmes -. This is the largest number of new virus species discovered in a single studywhich greatly expands our knowledge about the viruses that live among us. Finding so many new viruses at once is astonishing, and is just the tip of the iceberg, opening up a world of discoveries. “There are millions more to be discovered, and we can apply this same approach to identify bacteria and parasites.”
Although RNA viruses are commonly associated with human diseases, they are also found in extreme environments around the world and even can play key roles in global ecosystems. In this study they were found living in the atmosphere, hot springs and hydrothermal vents.
“The fact that extreme environments contain so many types of viruses is just another example of their phenomenal diversity and tenacity to live in the most hostile environments, which potentially gives us clues about how viruses emerged and other elemental life forms,” adds Holmes.
The researchers built a deep learning algorithm, LucaProt, to calculate huge amounts of genetic sequence data, including extensive virus genomes up to 47,250 nucleotides and complex genomic information to discover more than 160,000 viruses.
“The vast majority of these viruses had already been sequenced and They were in public databases, but they were so divergent that no one knew what they were. – Holmes confirms -. They made up what is often known as a ‘dark matter’ sequence. “Our AI method was able to organize and categorize all of this disparate information, shedding light on the meaning of this dark matter for the first time.”
The tool of AI was trained to calculate this “dark matter” and identify viruses based on sequences and the secondary structures of the protein that all RNA viruses use to replicate. This made it possible to significantly accelerate the discovery of viruses, which, if traditional methods were used, would take a long time.
“We used to rely on tedious bioinformatics processes for virus discovery,” says Mang Shi, co-author of the study, “which limited the diversity we could explore. Now, we have a much more efficient AI-based model that offers exceptional sensitivity and specificity and, at the same time, it allows us to delve much deeper into viral diversity. “We plan to apply this model in several applications.”
“LucaProt represents a significant integration of cutting-edge AI technology and virology, demonstrating that AI can perform biological exploration tasks effectively – concludes Zhao-Rong Li, co-author of the study -. This integration provides valuable information and stimulus for further decoding of biological sequences and deconstruction of biological systems from a new perspective. Also we will continue our research in the field of AI for virology”.
The obvious next step, according to the study, is to train this method to find even more examples of this diversity. But there are other future steps that are just as important. Once the new viruses were discovered, Their abilities to infect humans and also the time windows will be evaluated taking into account climate change or the invasion of wild areas where animals can carry these viruses and our interaction with them increases.