Pharmaceutical companies are increasingly using artificial intelligence (AI) systems to develop new treatments and medicines. British startup Exscientia and Japanese pharmaceutical firm Sumitomo Dainippon Pharma have used algorithms that sift through thousands of potential compounds to develop a medication that will soon be clinically tested on humans. (The drug is meant to treat obsessive-compulsive disorder.)
AI medicine context
The partnership between Exscientia and Sumitomo Dainippon is one example of pharmaceutical companies and AI startups partnering to leverage machine learning to discover patterns that might be too subtle or complex for humans to recognize. Furthermore, AI can be used to predict the reactions and behaviors of tiny molecules within chemical compounds that could produce desirable properties for medicines. For instance, Iktos, a company that leverages AI to discover new drugs, is applying its technology to a number of small-molecule discovery programs in collaboration with pharmaceutical giant Pfizer.
Another example is the partnership between AstraZeneca and BenevolentAI. The two companies are collaborating to discover new drugs for chronic kidney disease and idiopathic pulmonary fibrosis—a serious lung disease. German biotechnology company, Evotec, has also formed a partnership with Exscientia. Evotec has announced a phase one clinical trial on a new anticancer molecule, created in partnership with Exscientia.
By using data, machine learning, and AI, pharmaceutical companies and AI startups are pooling their expertise to discover new drugs faster than ever before. Algorithms may also identify molecules previously unknown to scientists, which could lead to new insights about diseases and how they can be treated or cured.
The use of AI can increase the speed at which new drugs can be developed and tested. For example, algorithms can sift through vast quantities of data on potential compounds, checking them against available databases of parameters in a fraction of the time it would take a battery of scientists. In the case of the Exscientia trial, a drug that typically would have taken about five years to be approved for a trial only took 12 months.
Applications for AI-powered pharmaceutical research
AI-powered pharmaceutical research has the potential to:
- Enhance and accelerate drug discovery and manufacturing.
- Facilitate the development of more effective and better-targeted drugs.
- Create more partnerships between pharmaceutical and AI companies to meet the urgent need for drugs that can treat many intractable and hereto incurable diseases.
- Produce more effective research results. Up until 2010, only ~10 percent of medicines in development ever reached the market. With the help of AI solutions, drug development may become more successful.
Questions to comment on
- Would you take a drug discovered and developed with the application of AI systems?
- What do you believe to be potential risks in using drugs and treatments developed by AI systems for human conditions?