Health
Published on February 22, 2020 |
by Steve Hanley
February 22, 2020 of Steve Hanley
Alexander Fleming first discovered penicillin in 1928. Although his work was based on the efforts of numerous researchers who preceded it, he was eventually awarded the Nobel Prize for medicine. At his acceptance speech, he warned that one day the bacteria could adapt to penicillin and make it less useful, and that’s exactly what happened. Today there are many drug-resistant bacteria that shake off the palliative effect of penicillin and many other antibiotics that have been widely prescribed by doctors, often with the encouragement of pharmaceutical companies.
Creating new drugs is a painstaking process that can be painfully slow. Developing new drugs can take years or sometimes decades, which is why drug prices are often extremely high. Someone has to pay for all the labs and technicians who research new drugs year after year. Many times, doctors don’t even know if the drugs they are working on are effective until they are far ahead in their research.
MIT researchers used a machine learning algorithm to identify a drug called halicin that kills many strains of bacteria. Alicin (top row) prevented the development of antibiotic resistance in E. coli, while ciprofloxacin (bottom row) did not. Image: courtesy of Collins Lab at MIT.
Artificial intelligence can help drastically reduce the time it takes to discover new drugs. MIT researchers say that using an AI algorithm, they identified a new antibiotic that kills many drug-resistant bacteria, according to a report by Science Daily. It has also proven effective in two studies in infected mice.
What makes this discovery even more remarkable is that a computer was able to do this in just three days. The algorithm is designed to identify potential antibiotics that kill bacteria using different mechanisms than those used by existing drugs, according to MIT news.
“We wanted to develop a platform that would allow us to harness the power of artificial intelligence to usher in a new era of antibiotic drug discovery,” says James Collins, professor of medical engineering and science. “Our approach has revealed this extraordinary molecule which is probably one of the most potent antibiotics that has been discovered.”
“We are facing a growing crisis around antibiotic resistance and this situation is generated both by an increasing number of pathogens that become resistant to existing antibiotics, and by an anemic pipeline in the biotechnology and pharmaceutical industries for new antibiotics,” says Collins.
The algorithm has also identified several other potential antibiotics that will undergo further testing. “The machine learning model can explore … large chemical spaces that can be prohibitive for traditional experimental approaches,” says Regina Barzilay, professor of electrical and computer engineering at MIT.
Until now, computer modeling was too imprecise to produce useful results, but the latest neural networks can learn to automatically express potential antibiotics in computer terms. The researchers designed their new model to look for chemical characteristics that make molecules effective in killing E. coli bacteria. They formed the model on around 2,500 molecules, including around 1,700 FDA approved drugs and a set of 800 natural products with different structures and a wide range of bio-activities.
Once the model was trained, researchers tested it on approximately 6,000 compounds. The model identified a molecule that had a strong antibacterial activity and a chemical structure different from any existing antibiotic. That molecule was called allicin, a name derived from HAL, the computer described in the Stanley Kubrick film 2001: A Space Odyssey.
[Extraneous side note #1: the first iPhone design was inspired by the black monolith shown in the opening sequence to that movie. Extraneous side note #2: it is believed Kubrick chose HAL because the letters all precede the initials IBM, which was the name of the dominant computer company in the world at the time of the movie. You’re welcome.]
The researchers tested against dozens of bacterial strains isolated from patients and grown in laboratory dishes and found that it was able to kill many who are resistant to treatment, including Clostridium difficile, Acinetobacter baumannii and Mycobacterium tuberculosis. The drug worked against every species tested, except for Pseudomonas aeruginosa, a difficult to treat lung pathogen.
MIT news he adds that alicin can kill bacteria by disrupting their ability to produce ATP, a molecule that cells use to store energy. Researchers think cells will have difficulty adapting to such a disruptive process. “When you are dealing with a molecule that is likely to associate with membrane components, a cell may not necessarily acquire a single mutation or a couple of mutations to change the chemistry of the outer membrane. Mutations like this tend to be much more complex than acquire evolutionarily, ”says Stokes.
After identifying alicin, the researchers used their model to select over 100 million molecules selected from the ZINC15 database, an online collection of around 1.5 billion chemical compounds. This screen, which lasted only three days, identified 23 candidates structurally different from existing and predicted non-toxic antibiotics for human cells.
In laboratory tests against five species of bacteria, the researchers found that eight of the molecules showed antibacterial activity and two were particularly potent. Researchers are now planning to further test these molecules and also to examine more of the ZINC15 database.
The researchers also plan to use their model to design new antibiotics and optimize existing molecules. For example, they could train the model to add features that could target a given antibiotic only on certain bacteria, preventing it from killing beneficial bacteria in a patient’s digestive tract.
So today scientists can do in three days what it normally takes months, years or even decades to accomplish. Imagine if that kind of research power could be applied to the search for antidotes on an overheated planet!
Follow CleanTechnica on Google News.
It will make you happy and help you live in peace for the rest of your life.
About the author
Steve Hanley Steve writes about the interface between technology and sustainability from his homes in Florida and Connecticut or anywhere else where Singularity could guide him. You can follow it chirping but not on any social media platform run by evil gentlemen like Facebook.