SOURCE / INDUSTRIES
Tencent makes foray into new drug R&D based on AI technologies
Published: Jul 09, 2020 09:48 PM

Aerial photo taken on May 19, 2019 shows the Tencent Guian Qixing Center (L) and the construction site of a Huawei Cloud data center in Gui'an New Area of southwest China's Guizhou Province. (Photo: Xinhua)



Chinese tech giant Tencent announced Thursday that it has launched a drug discovery platform driven by artificial intelligence (AI) technologies, aiming to reduce time spent on the research and development (R&D) of new drugs.

Called iDrug, the platform is committed to helping users significantly reduce the time and cost of finding potentially active drugs. Given industry experience, AI technology can shorten the new drug discovery cycle to six months to one year at the fastest speed, down from three to six years, the company said.

The platform, based on deep learning algorithms, can provide data base and cloud computing, covering five modules of the whole process of pre-clinical new drug development, which includes the finding and confirmation of the target, the key protein that eventually forms the disease.

Li Tianquan, co-founder of domestic healthcare big data platform yaozh.com, told the Global Times on Thursday that if the target is likened to a lock, it needs a key - a matched drug - to unlock it and thus cure the disease.

"AI can play a helpful role in the whole process by significantly reducing time and costs," said Li.

"Even though the technology's application in the pharmaceutical sector is still in its infancy, finding new drugs via AI represents the future trend of the industry, and several international drug giants have started exploring the sector," he noted.

Mark Ren, chief operating officer at Tencent, said in a greeting letter at the opening ceremony of the 2020 World Artificial Intelligence Conference (WAIC) on Thursday that the platform is currently trying to screen effective drugs against the coronavirus.

So far the iDrug platform has about 10 projects under development.