News

Using artificial intelligence to discover new treatments for superbugs

Machine learning is pointing researchers toward molecules that are structurally different from current antibiotics
Fabiola De Marchi
By Fabiola De Marchi
July 11, 2021

Antimicrobial resistance is an emerging threat to healthcare systems worldwide. As a consequence of the spread of drug-resistant bacteria, also called “superbugs,” medical treatments could become ineffective for an increasing number of people in the next years. To fix this huge problem, chemists are asked to find new effective antibiotics. 

Superbugs-445x297.jpg
Alexander Klepnev on Wikimedia Commons.

Drug discovery is an expensive and time-consuming process during which pharmaceutical chemists look for new candidate molecules to interact with a particular target protein or pathway causing the disease. Chemists screen large libraries of thousands to millions of molecules, looking for compounds with specific biological effects and low toxicity. However, these screenings are not very efficient: if chemical libraries don’t include molecules with enough structural diversity, chemists will fail to discover antibiotics with molecular structures different from the ones already tested in laboratories or clinical trials. 

Now machine learning is flanking chemoinformatics through innovative deep neural network approaches to find new drugs. An example of how this approach works can be seen in a recent study by James Collins and coworkers at MIT. First, researchers trained a neural network model to predict growth inhibition of Escherichia coli using a set of 2335 diverse molecules; then, they applied the optimized neural network model to screen large chemical libraries with more than 107 million molecules. 

Recent improvements in machine learning can speed up and lower the costs of drug discovery

They ended up with a list of candidate molecules structurally different from known antibiotics, and ranked them based on their predicted biological activity. Among those candidates, they found that halicin, a compound under investigation as a treatment for diabetes, displayed high efficacy against E. coli and a large spectrum of pathogens such as Acinetobacter baumanii, at the top list of resistant bacteria which urgently requires new antibiotics.

Research groups are currently developing similar deep learning approaches to find new compounds that could fight the COVID-19 virus. This suggests how recent improvements in machine learning can assist chemists’ work to speed up and lower the costs of the drug discovery process.

This story originally appeared on Massive Science, an editorial partner site that publishes science stories by scientists. Subscribe to their newsletter to get even more science sent straight to you.

Enjoy reading ASBMB Today?

Become a member to receive the print edition monthly and the digital edition weekly.

Learn more
Fabiola De Marchi
Fabiola De Marchi

Fabiola De Marchi is a science writer for Massive Science.

Get the latest from ASBMB Today

Enter your email address, and we’ll send you a weekly email with recent articles, interviews and more.

Latest in Science

Science highlights or most popular articles

Is there a mystery in the monkeypox antiviral?
News

Is there a mystery in the monkeypox antiviral?

May 24, 2022

As monkeypox becomes an international concern, interest grows in tecovirimat; this smallpox drug targets a structural protein that helps wrap the virus in a second lipid bilayer.

Researchers investigate self-regulation of an enzyme with critical cellular functions
News

Researchers investigate self-regulation of an enzyme with critical cellular functions

May 24, 2022

They found that one mechanism of CK1 activity, and thus one mechanism of regulation, is the self-phosphorylation of a conserved amino acid residue in its catalytic domain.

What is monkeypox?
Science Communication

What is monkeypox?

May 23, 2022

A microbiologist explains what’s known about this smallpox cousin.

A simple method to determine phase preference of proteins on live cell membranes
Journal News

A simple method to determine phase preference of proteins on live cell membranes

May 22, 2022

“The phase preference of molecules used to be difficult and time-consuming to establish. This new method, detected by chance, provides results in at most 15 minutes on live cells,” Thorsten Wohland said.

Study of Alzheimer’s marker prompts warning about serine supplements
News

Study of Alzheimer’s marker prompts warning about serine supplements

May 20, 2022

They have been advertised to improve memory and cognitive function.

Why is the 100-year-old BCG vaccine so broadly protective in newborns?
News

Why is the 100-year-old BCG vaccine so broadly protective in newborns?

May 19, 2022

Study finds changes in metabolite and lipid profiles, providing clues for designing future vaccines for newborns.