United States: A new publication selected probable new antibiotics in the global microbiome using the machine learning approach, something the study authors say is an innovation in the application of artificial intelligence in antibiotic resistance research.
More about the research
According to César de la Fuente, an author of the study and professor at the University of Pennsylvania, the details brought by the study helped scientists who used an algorithm to mine the “entirety of the microbial diversity that we have on earth – or a huge representation of that – and find almost 1m new molecules encoded or hidden within all that microbial dark matter”, as the Guardian reported.
De la Fuente leads the so-called Machine Biology Group under the main idea to harness computer capabilities to enhance biology and medicine advances.
Otherwise, said de la Fuente, scientists would have had to rely on ordinary approaches such as taking a sample of water or soil and looking for the molecules within it. It can be cumbersome because, even if one tries, microbes exist in the ocean, in the human gut, and practically in every corner of the world.
According to de la Fuente, “It would have taken many, many, many, many years to do that, but with an algorithm, we can sort through vast amounts of information, and it just speeds up the process,” the Guardian reported.
Growing antimicrobial resistance
The research is dire to public health, according to the author, for antimicrobial resistance caused more than 1.2 million people to die from cancer-related illness in 2019. It threatens to rise to 10 million deaths per year by 2050, as pointed out by the WHO.
According to de la Fuente, the study has produced the “greatest antibiotic discovery effort ever,” which has brought a momentous moment revealing the benefits of artificial intelligence for research, as he also acknowledged that bad actors could potentially “develop AI models to design toxins.”
Artificial intelligence and antibiotic research
Amid the recent debate over artificial intelligence’s benefits and drawbacks, de la Fuente noted that he introduced the technology to his work on antibiotics around ten years ago.
As de la Fuente added, “We have been able to just accelerate the discovery of antibiotics,” and, “So instead of having to wait five, six years to come up with one candidate, now, on the computer, we can, in just a few hours, come up with hundreds of thousands of candidates.”
Before the US Food and Drug Administration (FDA) grants approval for an antibiotic, it usually takes several years to go through tests in some laboratories and clinical trials. These various stages may require 10 to 20 years as one goes through continuous learning, risk-taking, the development of prototypes, testing, and market research.
For the purpose of this experiment, the researchers obtained all possible genomes and meta-genomes from any accessible database and searched for sequences potentially possessing antimicrobial properties.
For this study, the researchers collected genomes and meta-genomes stored in publicly available databases and looked for DNA snippets that could have antimicrobial activity.