Computational power is being leveraged to monitor clinical trials of drugs, vaccines, identify new compounds of medicinal value, etc.
The use of technology for a range of public-health purposes has, without doubt, proliferated after Covid-19. While over 50 nations—including India, Singapore, Canada, Israel, and Germany—have rolled out contact tracing apps, private sector tech companies have also come up with tech-led solutions to several Covid-19-related problems. But, the fact of note is, tech is even aiding better understanding of the disease; recent research that points to SARS-CoV-2 triggering a malfunction of the delicate interplay of bio-molecules that regulate vasodilation/constriction—which, in turn, could explain many symptoms of the disease—was made possible because of a supercomputer, Summit, crunching data gleaned from 30,000 genes from 17,000 genetic samples. While this is emerging research in the Covid-19 context—many more blanks need to be filled through deeper work—Summit’s data-crunching and the analysis it facilitated seem to confirm certain aspects of earlier research in the area. What is also interesting is that Summit has been used to identify 77 potential Covid treatments, based on the analysis of the data it processed.
Computational power is being leveraged to monitor clinical trials of drugs, vaccines, identify new compounds of medicinal value, etc. Indeed, Moderna, AstraZeneca and others used Medidata, from Dassault Systemes, which aided data management and analysis for their vaccine research. Medidata has also been used to create synthetic control arms, by augmenting placebo research with a virtual group, based on data gleaned from previous research; this not only improves the speed of research but also helps maintain trial quality and rigour. Another software from Dassault Systemes, Biovia, has been extensively used by institutions conducting vaccine research—based on past research data, it allows scientists to model viral protein behaviour and the evolution of epitopes (the part of the antigen that is recognised by the immune system). This helps researchers make inferences from past research to decide their own research strategies. Last month, IBM announced the launch of RoboRXN, which will help develop new drugs. The process is expected to cut the cost of drug development by a tenth as RoboRXN can also be used to predict and model molecule reactions during drug development. Relying on artificial intelligence, RoboRXN also learns from past mistakes. Earlier this year, a machine-learning algorithm at MIT discovered a powerful new antibiotic compound that proved effective against many antibiotic-resistant bacteria. The system is designed to screen millions of compounds to arrive at those that can kill bacteria in a manner that is different from what existing antibiotics follow. It isn’t difficult to imagine what this could do for the antibiotic pipeline that had dried up over the past two-three decades. The recent progress in data analysis, technology and AI development—and the ongoing R&D adding to this every day—will yield a rich dividend for science, especially medical science and physiology. To that end, countries investing in these areas—through both public sector and private players—stand to leap past the rest, something that India needs to keep in mind.