AI Drug Company Creating Covid-19 Therapy for Rapid Delivery

(Gerd Altmann, Pixabay)

24 May 2022. A company applying artificial intelligence to drug discovery says it designed a new oral drug for Covid-19 infections that can be quickly produced. Insilico Medicine says its yet-unnamed, preclinical, small-molecule therapy candidate acts against SARS-CoV-2 viruses responsible for Covid-19 and other coronaviruses, and can be quickly synthesized, reducing production time.

Insilico Medicine, based in New York, provides drug discovery and design as a service to biotechnology and pharmaceutical companies, but in recent months began discovering its own drug candidates and advancing them into clinical trials. The company’s technology uses deep learning, a class of A.I. algorithms that finds underlying relationship patterns, then builds those relationships into a drug discovery database. Insilico says its process applies an A.I. technique called generative adversarial networks that use two sets of algorithms to test each other, helping reveal and learn underlying relationships, while also optimizing data.

With these data relationships defined, says the company, the next step is identifying targets through another deep learning system that analyzes gene, transcription, protein, and clinical data sets. Insilico says it then applies its drug design technology to create therapeutic molecules from scratch, which meet predefined criteria and properties, such as chemical complexity, molecular shape, metabolic stability, and solubility. In February 2022, Science & Enterprise reported on the company’s first drug candidate for idiopathic pulmonary fibrosis, discovered and designed in 18 months, that began early-stage clinical trials.

Targeting replication enzymes

Insilico says it started investigating therapies for Covid-19 infections early in the pandemic. While many drug makers first sought to apply available anti-viral drugs to treating Covid-19, Insilico says its initial research showed drawbacks with many current anti-virals when applied to SARS-CoV-2 viruses, and chose instead to discover new therapeutic molecules. And its initial work also sought a different target from other drugs. Instead of addressing proteins on the surface of SARS-CoV-2 spikes, as most coronavirus vaccines and treatments up to now, Insilico targeted protease enzymes that cause the virus to replicate.

That investigation, says Insilico, revealed a 3C-like protease in SARS-CoV-2 and other coronaviruses as a prime target. The company says the 3C-like or 3CL protease in SARS-CoV-2 is active early in the viral replication process, thus blocking its effects can limit the virus from reproducing and spreading. “This molecule designed by AI has distinct pharmacophores from existing 3CL protease inhibitors,” says Insilico’s chief scientist Feng Ren in a company statement released through Cision, “and binds to the target protein in a unique, irreversible, covalent binding mode as demonstrated by a co-crystal structure.”

Using its design process, Insilico says it created a small-molecule oral drug candidate that inhibits the 3C-like protease in SARS-CoV-2 and other coronaviruses. The company says these molecules can be efficiently synthesized from two readily available raw materials, thus produced quickly. In addition, says Insilico, initial preclinical studies suggest the drug works effectively in low doses, and can be given alone instead of combined with other anti-viral drugs.

Ren adds Insilico Medicine is “committed to progress the molecule as fast as possible into clinical trials evaluating its usage in Covid-19 treatment.”

More from Science & Enterprise:

• Infographic – 100 Promising AI Companies
• AI-Aided Collaboration to Discover Neuro Disease Drugs
• Synthetic Bio Companies Partner on Therapeutic Microbes
• Trial to Test AI-Designed Cancer Antibody Therapy
• Preclinical Study Shows Immune Drug Prevents Covid-19 Death

We designed Science & Enterprise for busy readers including investors, researchers, entrepreneurs, and students. Except for a narrow cookies and privacy strip for first-time visitors, we have no pop-ups blocking the entire page, nor distracting animated GIF graphics. If you want to subscribe for daily email alerts, you can do that here, or find the link in the upper left-hand corner of the desktop page. The site is free, with no paywall. But, of course, donations are gratefully accepted.

*     *     *