How computational drug discovery is boosting health tech innovation
Image credit: Gajendra Bhati/Dreamstime
The devastating impact of Covid-19 has highlighted the urgent need for rapid identification of drugs that can be used to prevent or treat potentially fatal viruses and diseases. The use of new techniques in India illustrates the important part technology will play.
The process of drug discovery is extremely complex and requires time and resources to isolate the molecules capable of identifying the properties of a disease or virus. Health institutions are becoming increasingly dependent on the ability of computational drug discovery (CDD) to speed up and automate factors such as analytics and large-scale simulations for the trial process. To efficiently do this, drug developers are leveraging emerging technologies such as artificial intelligence and machine learning.
It has been a long wait for the pharmaceutical industry to find a technology that not only increases the effectiveness of the drug discovery and development process, but also addresses the traditional challenges of time and cost. Organisations embracing CDD are placing big bets on the technology while they adjust their vision for the future.
Across all sectors, AI has been one of the technologies at the forefront of recent digital-transformation initiatives, and the same is true of biotechnology. In fact, major drug developers are recognising its potential to accelerate the analysis of massive databases, while also saving on costs and promoting better decision making.
Big corporations such as AstraZeneca, Pfizer and Johnson & Johnson have made significant progress by leveraging artificial intelligence for genomics and engineering tools in pharmaceutical applications, helping improve target identification and validation. On average, new drugs take 10-15 years to reach the market, and half of that time is spent on clinical trials. AI’s ability to analyse a wealth of data in real time can help accelerate this part of the process by identifying diseases more clearly.
Similar time-saving outcomes can be enjoyed through machine learning (ML), a subfield of artificial intelligence. ML in drug discovery can use algorithms that recognise familiar patterns within sets of data that have been further classified. Deep learning (a subfield of ML) then engages artificial neural networks to mimic the transmission of electrical impulses in the human brain. This helps drug developers discover the effects of a drug molecule before commencing trials.
India is a perfect example of how CDD can operate effectively under intense pressure. The first case of coronavirus in India – the second most populated country in the world – was reported in the state of Kerala in January 2020. The pandemic made its way through India at a pace that has since staggered scientists.
During the height of the second wave, the severity of Covid-19 meant that hospitals were running out of oxygen and beds, and many people who had been taken ill were being turned away. While antiviral and immunosuppressive drugs like Remdesivir and Tocilizumab were being offered for severe cases of the virus, more needed to be done to accelerate trials and reduce the number of active cases and daily casualties.
Significant strides were made through initiatives like Tech Mahindra’s partnership with Reagene Biosciences. By May 2021, Tech Mahindra was working towards filing a patent for a drug molecule that can potentially attack coronavirus. Fast forward four months, and we are on track to start trials for a Covid drug by the end of the year.
Makers Lab, Tech Mahindra’s internal lab created to boost innovation, conducted the molecular docking analysis in-silico. Based on computational docking, machine learning and modelling studies, Tech Mahindra was able to fast-track the analysis process and shortlisted 17 drug molecules from a list of 8,000 FDA-approved drugs.
These 17 were tested in-vitro by partners Reagene Biosciences and Indras pvt ltd before three molecules that emerged successfully were assessed using a 3D bio-printed human vascular lung model to discover which was most effective in combating the symptoms of coronavirus. This is a significant step forward that can speed up the drug discovery mechanism in biological computation.
Considering the spread and severity of certain diseases, such as Covid-19 in India, brings to light just how essential innovation is within the drug-discovery process. We live in an era where waiting 10-15 years for an FDA-approved drug is no longer a viable option. Health tech will continue to innovate, and we are only at the beginning of discovering the life-changing effects of emerging technologies such as machine learning. And with quantum computing knocking at the door, the health sector is poised to experience a significant boost to the ability of AI techniques to solve some of these latent challenges.
Nikhil Malhotra is chief innovation officer at Tech Mahindra.
Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.