More and more pharmaceutical companies are turning to computer-aided tools and methods to assist in discovery efforts, helping to solve complex problems in drug design and development.
Big Data, machine learning and artificial intelligence — more and more pharmaceutical companies are turning to computer-aided tools and methods to assist in discovery efforts, helping to solve complex problems in drug design and development.
As competition and the desire to be first-to-market increases, AI can give companies an advantage for the next big breakthrough. It can also potentially reduce development costs by streamlining the process of researching a drug and putting it through a clinical trial. At the same time, it can also target causes of adverse reactions that lead to a drug being removed from market.
With the cost of launching a new drug estimated at $2.6 billion pre-market and an additional $312 million for post-approval research and development, pharmaceutical companies are wise to use every means to manage their resources. So how can these tools help you? If your company isn’t using these tools, should you be? What are the benefits? Data shows that these tools are worthwhile, but they need to be used properly.
Deep Learning Produces New Breakthrough
A recent paper published in Nature Biotechnology cites research that used deep learning to generate novel small molecules for a protein target. “The traditional drug discovery starts with the testing of thousands of small molecules in order to get to just a few lead-like molecules and only about one in 10 of these molecules pass clinical trials in human patients,” notes SciTechDaily. “Even a slight improvement in the time it takes to discover new drugs or in the probability of success results in significant savings and public benefit.”
The efficiencies realized by this discovery can reduce the time and cost necessary to qualify a new drug for development. Machine learning and AI have the potential to resolve several challenges, ensuring that a safe and effective product makes it to market.
Diagnosing Cause of Adverse Reaction
AI can also help determine the cause of adverse reactions in drugs that are already on the market. More than 450 medicines worldwide have been withdrawn from the market post-approval as a result of adverse reactions. AI can help discover the metabolism of compounds by organs and discover if it is toxic or not before it goes through clinical trials and/or market approval. In addition to achieving better patient outcomes, this represents a huge cost savings for drug companies.
In 2018, for instance, AI was used to discover the two-step process in which an approved drug was causing liver toxicity — a process that was difficult to determine experimentally.
“Given enough data, machine learning algorithms can identify patterns, and then use those patterns to make predictions or classify new data much faster than any human,” notes a recent article in The Scientist. For the complex data required to capture patterns, AI can offer a distinct edge.
Once you have trained a machine learning algorithm on a data set of an existing molecule, it can offer predictive information on whether or not new molecules are also toxic.
Machine learning algorithms could also predict how a candidate molecule will respond to different physical and chemical environments, helping to show how that molecule might behave in the human body.
Machine learning and AI are powerful tools that, if used properly, have the ability to transform drug discovery, which could correlate to increased speed in identifying drug candidates and get important drugs to market quicker. It could also save significant clinical trial costs by moving forward candidates that may be more successful and less toxic. Further, AI could help identify targets that are safer for patients and avoid post-market recalls. If you’re not using these tools, it might be a good time to consider them.