Artificial Intelligence used to design brand new viruses
Scientists made 16 successful viruses that had their genetic code designed by artificial intelligence.
US researchers have employed artificial intelligence (AI) to design entirely new viruses that can replicate in laboratory conditions, according to a report. This marks the first instance where AI successfully created a complete virus genome. The 16 novel viruses, engineered to infect bacteria without posing a threat to humans, are poised to unlock new avenues in treating disease.
However, experts caution that AI-designed viruses present urgent safety and security concerns. Although AI has been utilized for designing new antibiotics, creating a viable virus from scratch represents a markedly more complex challenge. Brian Hie, an assistant professor at Stanford University, expressed amazement at the groundbreaking nature of this achievement.
The technology functions akin to large language models, like ChatGPT, which predict text sequences. In this scenario, AI models known as Evo1 and Evo2 predicted biological sequences instead of words. The models, trained on genetic codes from various life forms, were refined to produce bacteriophages, viruses that target specific bacterial species.
Out of 302 promising designs, 16 were synthesized and proved effective against E. coli bacteria. The successful creation of these phages in the lab has sparked excitement and applause among the researchers. These advances in synthetic biology could lead to novel treatments for antibiotic-resistant infections, but they also raise significant biosafety and biosecurity concerns.
The researchers took precautions to ensure safety, such as excluding viruses capable of infecting complex organisms from the training data and conducting the research on phages rather than human viruses. The genetic code of the phages is relatively small, consisting of around 5,400 base pairs, compared to the smallest living cell genome of about 500,000 base pairs and the human genome's three billion base pairs.
Prof Marc Güell from Spain's synthetic biology lab highlighted the study as a significant turning point, as it marks the beginning of designing biology on a computer, opening up possibilities for tackling major global challenges.
Written by urgent.news from BBC Health's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
