SK Biopharma said Monday that it has launched a drug design platform based on artificial intelligence that can reduce the research cost and speed up the development of a new drug.
The company based the platform on its extensive research data and researchers' experience that it accumulated over the past 20 years, and has been collaborating with SK C&C to complete AI algorithm optimized for new drug development.
SK Biopharma expects that the platform will reduce the development period and the failure rate of new drugs, which, until now, took an average of more than 10 years.
The AI system consists of three technologies – AI model that predicts drug properties and designs the experiment, a compound data storage and an AI data storage.
The AI model, which uses SK C&C’s machine and deep learning technologies, uses a “drug characteristic prediction” model that can identify the absorption, distribution, metabolism, excretion and toxicity of a drug.
The model also includes a drug experiment design model, which uses the results as mentioned earlier to design and propose new compounds by grasping the hidden patterns and attributes of the drug in the data.
Although Korea already has various platforms that can predict drug properties, the company believes that its platform is the only system that designs new compounds that are not currently available.
Regarding compound data storage, the platform is a collection of internal and external data containing preliminary information and patent information of the compound that goes through the company’s machine learning algorithm and stores itself so that researchers can search and use the data at a later period. The company plans to mount the latest learning data provided by the compound data storage onto its AI model platform.
“The developed AI platform, which is a unique platform that enables designing the model that was in the previous not available, has opened the door for a new drug development process,” said Maeng Chul-young, director of SK Biopharma’s digital healthcare taskforce.
The company expects that the platform will accelerate the process of finding promising new drug candidates, which will dramatically shorten the time and cost of the development, he added.