AI Tool to Classify COVID-19 Genome Signature Developed by Group of Scientists; Indian-Origin Scientist Gurjit Randhawa, Part of Team, Speaks to LatestLY Exclusively

SARS-CoV-2, which was first reported from China’s Wuhan City, has spread across most of the countries of the world. A team of scientists has developed an artificial intelligence tool for the elucidation of the taxonomic classification and evidence supporting a possible origin of the virus. Indian-origin scientist Gurjit Randhawa is part of an interdisciplinary joint research project of the University of Waterloo and the University of Western Ontari. The research team developed an AI-based method to identify an underlying genomic signature for 29 different RNA sequences of the novel coronavirus.

The software developed by Gurjit Randhawa’s team classifies the pathogen in minutes. According to Randhawa, the open-source software has a user-friendly interactive interface. It is a standalone software that is publicly available. Randhawa and his team are now working to make the tool available on the web for everyone. Randhawa’s teammate Maximillian Soltysiak looked after the aspects of biology while carrying out the research. He said that once the biological identity of the virus is known, it helps to form strategies for further study.  The other members of the team are - Hadi El Roz, Camila PE de Souza, Kathleen A Hill and Lila Kari. US Scientists Identify New Mutated Version of Coronavirus, Claim 'New Strain More Contagious'.

This pattern-based search yields results considerably faster compared to traditional methods. According to Randhawa, the study generates accurate results, and it does not only apply to pathogens but also other species. The early classification of the virus could help scientists develop a vaccine in relatively less time. It could also serve as a guideline for authorities to understand the potential threat from a particular pathogen, potentially indicating the need for pro-active measures to contain the spread. In an exclusive interview with LatestLY, Gurjit Randhawa and Maximillian Soltysiak explained the methodology and the utility of the research. Notably, Randhawa is also pursuing Ph.D from the University of Ontario. Coronavirus Outbreak: Conspiracy Theories Floating on Facebook, Twitter, WhatsApp About Deadly Virus Spread.

Here Are Excerpts of the Interview:

Could you elaborate on how taxonomic classifications of the virus will help in the treatment/vaccine?

Taxonomic classification of the virus and for any other emergent pathogen helps in the very early stages when the identity of the organism is not known. Classifying the unknown organism quickly provides a wealth of biological information and experiential knowledge based on what we already know about the closest match.  That species match can provide insight into the route of infection, spread, severity, and more which can be critical for developing a strategy for containment and developing vaccines.  Scientists can use that information as a start to learn more about the biology of the new virus or new pathogen.

How was AI useful in the research process? And what other ways is AI helping medical professionals/first responders in treating the COVID-19 patients?

For us, machine learning is a crucial component of our methodology. AI made it possible to establish a method that can quickly predict the taxonomic labels of any unknown sequence by comparing it against the thousands of known sequences. AI is being used for tasks ranging from taxonomic classification to structure prediction, drug discovery, and spread prediction studies.

We are getting reports about the virus mutating in some instances. How will the AI technology help in classifying the mutations?

Yes, a few mutations have been identified and used for tracking the spread of the virus between regions. Numerous studies have also observed the multiple clades of the COVID-19 virus.  AI can be used in quickly identifying the possible virus sub-types and tracking how the virus evolves. As more data becomes available, AI supervised learning algorithms could learn this information and could play a role in determining if these mutations alter the viral functions.

Your research supports the hypothesis of a possible bat origin. How different is it from the previous strains of coronavirus, which also have their roots in bats?

We have observed the data available at the time of this study and found the closeness of COVID-19 virus sequences to the RaTG13 bat Betacoronavirus sequence. The same has been observed by several other research groups. To be sure of any origin, more data is required, at least sequences from all the species which were being sold at the Wuhan seafood market (claimed to be ground zero). Bat is a host to many types of known coronaviruses. While our data supports a bat origin hypothesis for the virus, whether or not the virus crossed over into humans directly from bats or another intermediary species requires additional research, especially from samples taken from the wet market of proposed origin.

Can this type of genome analysis be implemented for other viruses from other animals and bats to learn more about the virus's origin?

Yes, all we need is the raw sequences. We have used this method in the past for many other applications, including the taxonomic classification of other organisms, virus sub-type classification, etc.

The question of the origin of the virus is very challenging to answer.  The first human infection happened at an unknown time and place.  We lack the direct sequence evidence from that infection event.

While our data support a bat origin hypothesis for the virus, whether or not the virus crossed over into humans directly from bats or another intermediary species requires additional research, especially from samples associated with the proposed origin.

Would you completely back China's wet market theory on coronavirus origin? Or is there any base to the lab-made virus conspiracy theories?

The right answer is that we cannot comment on this and cannot speculate, as this is not the topic of our research and not our area of scientific expertise.

How would the new tool expedite the development of a vaccine to combat coronavirus?

Our analyses are very relevant early on in virus discovery prior to its spread to a pandemic.  Our tool is a valuable resource for within-minutes identification of accurate taxonomic information, and this would make it a crucial tool for strategic planning and research for future pandemics.

Was there any insight found on whether the SARS-CoV-2 (COVID-19 virus) can evolve and lead to the second wave of the pandemic?

This is still unknown, so we should stay away from speculations and wait for a few months to have more information. We should listen to the advice of virologists and epidemiologists on this as it is their area of expertise.

What is the most conservative and optimistic estimate for vaccine development period?

This is not our area of expertise. We appreciate the effort of researchers working on vaccine development and hope for their early success.

What are your views about India's response to Coronavirus and How It can improve?

I do not have expertise in epidemiology, and I cannot comment on this scientifically, but my opinion as a lay-person is that the Indian government did well with closing the flight operations and enforcing lockdown at the right time. In my opinion, this was the only way for densely populated countries. Looking at the numbers, India is in a much better state considering the huge population. We should follow the advice of the epidemiologists and doctors on what measures are best at containing the outbreak.

The number of positive coronavirus cases has crossed 3.7 million until now. The deadly virus also claimed more than 250,000 lives globally. The United States is the most affected country in the world where the number of confirmed cases crossed one million, while close to 68,000 people also lost their lives.

(The above story first appeared on LatestLY on May 06, 2020 02:55 AM IST. For more news and updates on politics, world, sports, entertainment and lifestyle, log on to our website latestly.com).