South Korean pharmaceutical giant Daewoong Pharmaceutical said Monday that it has completed a new drug development system that secures drug candidates based on artificial intelligence technology.

It created a collection of 800 million preprocessed molecular models, called Daewoong Advanced Virtual Database, or DAVID, and a separate AI system, called Daewoong AI System, or DAVIS, to nurture new drug candidates throughout the preclinical, clinical and commercialization phases.

Daewoong said the extensive database is the outcome of more than 40 years of research.

For the AI-tailored database, the researchers went through essential preprocessing to separate and remove unnecessary information from complex compound structures, as publicly available compound open source data worldwide was unsuitable for AI drug development.

The company also unveiled the AI-based Virtual Screening tool to detect active substances targeting specific proteins based on 3D modeling technology, enabling the rapid discovery of new active substances with similar chemical characteristics eligible for patents.

Daewoong said the utilization of the AIVS tool has already yielded promising results, particularly in areas such as obesity, diabetes and anti-cancer drugs. Moreover, significant advancements have been made in developing protein degraders, drastically reducing trial-and-error processes through simultaneous antibody design and stability evaluation.

Meanwhile, applying AI technology could reduce the drug development period to seven years and the cost to approximately 600 billion won ($450 million), a drastic innovation from an average of a 15-year drug development period and development costs of 2 trillion to 3 trillion won, as reported by the National Institute of Food and Drug Safety Evaluation.

"When researchers and AI grow together, raising 'insights' together, we can take a step closer to the success of drug development,” said Park Joon-seok, head of the Drug Discovery Center at Daewoong.