On September 19, HUAWEI CONNECT 2022 kicked off in Bangkok, Thailand. At the event, Ken Hu, Rotating and Acting Chairman of Huawei, delivered a keynote titled “Unleash Digital,” where he talked about how the company is helping organizations make the most of cloud technology to leapfrog their developments.
During his speech, Hu spoke about how the First Affiliated Hospital of the Medical School at Xi’an Jiaotong University used AI to expedite pharmaceutical R&D and new drug discovery.
In the pharmaceutical industry, from development to approval, an average of over $1 billion over 10 years is often needed to put a new drug on the market. While developing antibiotics, it has been noted that resistant bacteria are sometimes discovered even before newly developed antibiotics finish clinical trials.
By using an AI-assisted drug design service powered by Huawei Cloud’s Pangu Drug Molecule Model, Professor Liu Bing of Xi’an Jiaotong University and his team developed a new broad-spectrum antimicrobial drug in just one month, slashing R&D costs by 70% in the process.
A major challenge in new drug discovery lies in the screening of hundreds of millions of existing drug molecules. Traditionally, drug screening has been performed by experts in labs, which is costly, slow, and has a high rate of failure.
The Huawei Cloud Pangu Drug Molecule Model has been trained using the data of 1.7 billion drug-life molecules, and can predict the physicochemical properties of drug compounds and score them based on their similarities. Researchers can then do targeted experiments to verify drug compounds that have the highest scores.
Moreover, the Pangu Drug Molecule Model’s molecular structure optimizer can be used to optimize the structure of lead compounds, minimizing the potential side effects of the new drugs on normal human cells.
For drug discovery, many pharmaceutical companies have partnered with AI vendors or start-ups to take advantage of their technology and expertise. A recent analysis by GlobalData identified almost 100 partnerships between AI vendors and large pharma companies for drug discovery since 2015, with increasing numbers witnessed in recent years.