Machine Learning: An Answer to Coronavirus

by April 1, 2020 0 comments

Coronavirus

We have seen terrible diseases get cured by modern sciences in movies or fascinated over the idea of fighting off the spread of illness in VR or Simulation games. However, the coronavirus is a reality against which human civilization is leaving no tables unturned to win this game. Here our choice of weapons is AI-based Machine Learning and Big Data Analytics.

In the healthcare sector, Machine Learning can be viewed as a resource that has the scope to process huge datasets beyond the capability of human minds and the inferred insights help physicians in planning and providing care to get satisfactory treatment. Researchers are looking to discover potential treatment solutions by investigating the already approved drug database and new compounds. While the traditional methods of drug discovery and development cost both time and money, experts rely on Computer-Aided Drug Design (CADD) as a method for developing candidate drugs for the treatment of many types of diseases. It is a computational approach to discover, develop, enhance and analyze drugs and similar biologically active molecules. This approach entails a vast amount of computational methodologies like virtual screening, virtual library design, lead optimization, de novo design and is used for rapid assessment of chemical libraries about compounds to guide and speed up the early-stage development of the new active biological compound.

Recently, two graduates of the Data Science Institute (DSI) at Columbia University are using this tech to quickly discover treatments for the novel coronavirus. Andrew Satz and Brett Averso (CEO and CTO respectively) of a 12-member startup called EVQLV claim that conducting antibody discovery in a lab can take years, an average of five years, and half a billion dollars to find high affinity and optimize antibodies.

“What our algorithms do is reduce the likelihood of drug-discovery failure in the lab,” Satz informed the institute newsgroup. “We fail on the computer as much as possible to reduce the possibility of downstream failure in the laboratory. And that shaves a significant amount of time from laborious and time-consuming work.”

Now EVQLV is working in collaboration with Immunoprecise Antibodies (IPA), a company focused on the discovery of therapeutic antibodies. Together they submitted their first panel of candidate therapeutic antibody sequences, comprised of DNA sequences encoding for potentially therapeutic antibodies against the new coronavirus, SARS-CoV-2.

“As new insights into SARS-CoV-2 are captured by researchers around the world, we are applying these insights in an attempt to generate an efficacious and sustainable therapy. The submission of the first panel of antibody sequences by EVQLV is an important milestone,” stated Dr. Jennifer Bath, CEO of IPA.

Although EVQLV has significantly minimized the initial phase of drug discovery, given the international urgency researchers are optimistic that possible treatment will be available by year-end.

While the stats about people tested positive keep fluctuating daily, it is only a matter of time till we find a bulletproof cure against this gruesome virus that has already claimed nearly 40k lives on a global scale. Till then stay indoors, practice clean habits, wait patiently till Machine Learning saves the day.

No Comments so far

Jump into a conversation

No Comments Yet!

You can be the one to start a conversation.

Your data will be safe!Your e-mail address will not be published. Also other data will not be shared with third person.