An AI identifies a powerful antibiotic to fight superbugs

• Read original article here

We have seen some of the most promising use cases of the Artificial intelligence system in the healthcare field recently. In one such example, Google’s AI lab DeepMind employed the futuristic tech to accurately detect the disease in Breast Cancer patients. Bacteria has increasingly become resistant to antibiotic drugs over the years as infections have become more dangerous. This has been amply manifested in the recent COVID-19 outbreak.

Efforts seem to be gathering pace to find a vaccination or cure for the disease which has rocked the World. A U.S based startup, Insilico Medicine has announced that is using Artificial intelligence to identify molecules that can expedite the formation of treatment for the virus.

Using AI is the right approach as we race against time to find a cure against such superbugs. Drug discovery is a painstakingly arduous task and requires analyzing copious amounts of data followed by animal trials & human trails before even being considered by the authorities for commercial use approval. Needless to say, the whole process takes years.

Unfortunately, in fighting viruses like COVID-19, which have the potential of becoming pandemics, we don’t have that much time. The good news is, however, that this is exactly the kind of job that AI excels at. In one such study, researchers at MIT & Harvard University trained a machine learning model on around 2,500 molecules, including existing FDA-approved drugs and other natural products.

After the AI system was trained on this extensive set of molecule data, it was made to scan a library of about 6,000 drug compounds to find the best candidates having the best antibacterial properties. One such molecule was named halicin — after the AI system HAL from the movie 2001: A Space Odyssey.

Halicin, which has now come across as a strong antibiotic agent was previously studied as a potential diabetes drug. The lab tests on all antibiotic-resistance bacteria — Clostridium difficile, Acinetobacter baumannii, and Mycobacterium tuberculosis,proved that the molecule successfully terminated these infections. The only exception was the bacteria Pseudomonas aeruginosa —which often infects the urinary or respiratory tract.

The molecule showed other impressive results as well. An ointment containing halicin was applied to mice infected with a strain of A. baumannii —a bacterial infectionresistant to all known types of antibiotics. This time the infections were completely cleared within 24 hours.

In the subsequent tests, the researchers found evidence that the drug works by disrupting the bacteria’s ability to maintain an electrochemical gradient on their outer membranes. Since the gradient is used by cells to store energy, a breakdown means that the bacteria cells die. The team believes it would be difficult for the bacteria to develop resistance to this killing mechanism.

Other common bacteria like E.coli,which is known to develop resistance to the antibiotic ciprofloxacin within one to three days, while becoming 200 times stronger after 30 days against the drug, was found to develop no resistance to halicin during the same trial period.

The next phase of the research will focus on conducting human trials. The AI system also shortlisted 23 other antibiotic candidates — two of which were found to be especially strong. Future testing would involve conducting trials on these as well.

The research was published in the journal Cell.