How Artificial Intelligence Is Helping Pharmaceuticals Develop Drugs

Theoretically, the process of drug development is a compact five-step process. However, in reality, it takes an average of 12 years, costing companies more than a billion dollars for a drug to travel from the laboratory to a pharmacy.

This long, complex, and expensive journey is challenging but it offers an opportunity to bring in an impactful change. This is why more and more pharmaceutical companies are exploring advanced technologies such as artificial intelligence, machine learning, big data and cloud computing to streamline the process of drug discovery to drastically reduce both time and costs. Here is an overview.

Drug Discovery

Drug discovery and development is the first step of the five-step process for drug development established by the FDA. It is followed by preclinical research, clinical research, FDA drug review and finally FDA post-market drug safety monitoring.

The California Biomedical Research Association estimates that, “Of all the tens of thousands of new drug compounds that begin the research process on the laboratory benchtop, only about five in 5,000 of the drugs that begin preclinical testing (animal trials) ever make it to human testing. Only one of these five is ever approved for human usage.”

In 2016, The Tufts Center for Study for Drug Development assessed the cost that goes into each successful drug at around $2.6 billion.

The conventional process of drug discovery starts with scientists beginning to identify and shortlist the chemical compounds from a humongous amount of data and images that could be tested for a targeted disease.

This extensive and time-consuming process is where machine learning can be applied to speed things up. The drug discovery market is projected to reach $85.8 billion in 2022, growing at a compound annual growth rate of 9.4% from 2017-2022.

Industry Collaborations

Some of the biggest names from the pharmaceutical world are working with technology giants and start-ups to shorten the time to identify new drugs and re-purpose current drugs for better healthcare.

Merck (MRK) was a pioneer among pharmaceutical giants to venture into the space of AI. Back in 2012, Merck partnered with Numerate to utilize its in silico drug design technology proprietary algorithms and cloud computing to generate new drug leads for an undisclosed cardiovascular disease target. Merck is a part of Atomwise which uses “a combination of patented structure based convolutional neural network for drug discovery, and enormous amounts of data.”

Since 2016, Pfizer (PFE) is leveraging IBM Watson (IBM) for drug discovery and support the identification of new drug targets, combination therapies for study, and patient selection strategies in immuno-oncology--an approach to cancer treatment that uses the body's immune system to help fight cancer.

GlaxoSmithKline (GSK) has partnered with several AI-driven companies to work on drug design and discovery. This year, GSK and Cloud Pharmaceuticals, Inc., joined hands for drug discovery wherein Cloud will design novel small-molecule agents to GSK specified targets. In July last year, GSK and Exscientia entered into a pre-clinical collaboration of £33 million to discover novel and selective small molecules for up to 10 disease-related targets, nominated by GSK across multiple therapeutic areas by leveraging Exscientia’s AI platform.

Leading the innovation at Johnson & Johnson (JNJ) is Janssen Research & Development which tackles research and development of new drug treatments and therapies. Janssen has been working on combining machine learning for drug discovery and based on its studies, it concludes that, “AI method to be up to 250 times more efficient than the traditional method of drug discovery.”

Last year, Genentech, a subsidiary of Roche (RHHBY) collaborated to GNS Healthcare (GNS) leverage GNS Reverse Engineering and Forward Simulation (GNS REFS) causal machine learning and simulation platform to accelerate cancer drug development.

Meanwhile, Novartis (NVS) and Intel (INTC) are using deep neural networks (DNN) to accelerate high content screening of cellular phenotypes, a critical element of early drug discovery. This year, representatives from both companies have shown “more than 20 times improvement in the time to process a dataset of 10K images for training” by shrinking it from 11 hours to 31 minutes.

In a ground-breaking development, Imperial College London, Singapore’s Duke-NUS Medical School and UCB, a Belgium-based pharmaceutical company have discovered a new drug target and a whole new approach—Causal Reasoning Analytical Framework for Target (CRAFT) discovery—that can discover and validate a potential new anti-epileptic drug in less than two years.

While advanced technologies are being explored for the healthcare industry as a whole, the potential that these technologies hold for drug discovery is of great interest to pharmaceutical giants that invest billions on cutting edge research and patents, every year. The potential of AI, big data, machine learning and cloud computing is set to streamline the whole drug discovery process, making it faster, cheaper and more efficient.

The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.

The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.

More Related Articles

Info icon

This data feed is not available at this time.

Sign up for the TradeTalks newsletter to receive your weekly dose of trading news, trends and education. Delivered Wednesdays.