Making use of the latest technology invented by mankind for the treatment of deadly diseases like COVID-19, scientists around the world have now discovered certain hundreds of drugs that can play a pivotal role in eliminating the coronavirus from the body of a patient infected by it.
Machine Learning (ML) has been incorporated into identifying these potential drugs, one of which is India-based.
The main author of the study, Anandasankar Ray belonging to the University of California, Riverside of the US said, “We have developed a drug discovery pipeline that identified several candidates.”
The pipeline Mr. Ray is referring to is a kind of computational strategy made useful with the assistance of AI, the technology employed worldwide in various industries. This is an algorithm or procedure which holds the capability to predict and improvise on the activities pertaining to the drugs through trial and error.
The study published in the journal called Heliyon says that a proper vaccine for the COVID-19 cannot be available in the near future as it months away. Moreover, this is still an uncertain timeline.
Mr. Ray said, “As a result, drug candidate pipelines, such as the one we developed, are extremely important to pursue as a first step toward the systematic discovery of new drugs for treating Covid-19.”
The drugs which are currently present in the market are FDA approved and act on one or more human proteins, vital for the access and replication of the virus in the body, have been made the top priority to recondition forming newer and more potent drugs for treating COVID-19.
“The demand is high for additional drugs or small molecules that can interfere with both entry and replication of SARS-CoV-2 in the body. Our drug discovery pipeline can help,” he added.
The Group of already discovered ligands for 65 human proteins with a known interaction with SARS-CoV-2 proteins formed the samples for landing at the result.
Every human protein was then tagged with an ML model to help the research team to design a record for the chemicals with structures that were expected to be capable of interacting with the 65 protein targets.
Along with this, the important step of safety was also tested to deem it fit for further continuation. The machine learning models were used for screening over 10 million commercially available small molecules from a database consisting of 200 million chemicals and identified the most potent ones for the 65 human proteins that interact with SARS-CoV-2 proteins.
Continuing the on COVID-19 study, the scientists found compounds from the FDA approved potent ones like the drugs and substituents used in food.
The ML models were further employed for computing the toxicity levels of the drugs to discard the toxic ones. The move aided in highlighting and working on the chemicals that seemed to interact with the SARS-CoV-2 targets.
Through this study, the drugs which had the most potency for treating the disease, or more so the contenders with substantial activity against a single human protein target could be identified along with some chemicals which had the ability to inhibit one than two human protein targets.