How artificial intelligence is influencing drug discovery

Posted Apr 15, 2018 by Tim Sandle
The drive for new medications addresses patient need and is the motor that keeps the pharmaceutical sector turning. To accelerate the drug discovery process, pharmaceutical organizations are turning to artificial intelligence.
Testing in a laboratory
Testing in a laboratory
Both major pharmaceutical companies and startups are applying artificial intelligence in drug discovery. Often the bigger players are linking up and forming partnerships with AI startups.
The drug discovery process is lengthy. Typically, it can take five years from a proposal, based on laboratory experiments, to see a new medicine hitting the market. The major gateposts are experimenting with different active ingredients; running clinical trials; and seeking regulatory approval from global regulatory, like the U.S. Food and Drug Administration.
To a degree, modern drug discovery is faster than before. The sequencing of the human genome, for example, has led to a faster process. This milestone in biological science allowed for rapid cloning and synthesis of large quantities of purified proteins, enabling high throughput screening of large compounds libraries against isolated biological targets.
Nevertheless, the desire for faster concept-to-market has led to pharmaceutical companies investing in artificial intelligence. This investment can be seen with emerging companies and the bigger players.
Start-up examples
Examples of start-ups include the company BioSymetrics, which deploys AI to process raw phenotypic, imaging, drug, and genomic data sets. This process allows scientists to add machine learning capabilities into new compound discoveries.
A second start-up making inroads is the firm Helix, which is using AI to respond to verbal questions and requests in the laboratory setting. This process permits researchers to raise efficiency levels and to keep up-to-date with relevant new research. There is also an added bonus in terms of using the platform to manage inventory.
A third start-up of note is Mozi, which has designed AI to identify patterns in biomedical data and, from this, infer hypotheses for investigation. Here scientists can analyse multiple datasets in the context of global biomedical knowledge. This process is particularly useful for personalized medicine initiatives.
Major players
In terms of major players, AstraZeneca and Berg Health have entered into a partnership designed to discover new therapeutic targets for neurological diseases like Parkinson's. In addition AstraZeneca has announced a collaboration with the company Alibaba to apply technology including artificial intelligence to patient diagnosis and treatment
These examples tally with a TechEmergence study on pharmaceutical sector executives. This showed that 50 percent anticipate broad scale AI adoption by 2025 and around half of the participants anticipate that initial AI applications will target “chronic conditions.”