The process of drug development is long and complex. New drugs go through animal trials, followed by human trials, a process that may take years. Furthermore, patients in trials can experience unforeseeable side-effects. After the trial is successful, it must be approved by a relevant regulatory agency (ie: the US Food and Drugs Administration, FDA).

Before outlining the benefits of AI, we must understand how drugs are developed. The first step is identifying the biological target (ie- cancer cells). After which a lead molecule is identified. This is a chemical compound that has potential to lead to a development of a new drug to treat a disease.This is used as a starting point for chemical modifications in order to synthesize a drug with the maximum therapeutic benefit and minimal potential for harm.

Many companies are currently collecting data and increasing their databases, in order to reduce the time taken for development and to reduce the cost to do so. In 2016,  the company Cambridge Cancer Genomics (CCG) started a project that aims to collect as much genomic data as possible.

The more clinical and genomic data oncologists have, the smarter decisions they can make about drug usage in any given circumstance.

These such companies understand that AI can save the cost in developing drugs. AI can be incorporated into various drug development stages, such as designing new drugs after incorporating and analysing datasets. For identified lead molecules, AI can predict its reaction with the biological target in silico, in a computer simulation. This helps scientists to eliminate poor candidates and increase the chances of obtaining and using good candidates, reducing the chances of the drug failing during clinical trials. 

For every drug developed, there is a specific synthetic route. The AI is able to predict the drug’s synthetic route based on data algorithms and comparing data from large databases. This allows scientists to assess how easy it is to synthesize a drug. It allows scientists to plan the development of the drug more efficiently. 

For more information, please read the article “Artificial Intelligence in Drug Development: Present Status and Future Prospects” published by Drug Discovery Today (Volume 4, Issue 3, Pages 773 to 780) here: https://www.sciencedirect.com/science/article/pii/S1359644618300916#bbib0240

Thumbnail credit: technologynetworks.com