[Edaily Reporter Shin Yeong-bin ] LG CNS (#LGCNES) announced on the 12th that it has completed the construction of an AI-based new drug development platform for the Dong-A Socio Group in collaboration with DAI, the group’s IT subsidiary.

The platform, developed by the two companies over approximately six months, integrates new drug research data and uses AI to support key research processes, from candidate compound discovery to validation. A key feature is its ability to continuously link and learn from AI prediction results and actual experimental data, thereby improving predictive performance as research progresses.

New drug development typically takes 10 to 15 years and incurs enormous costs, as it requires the repeated validation of numerous candidate compounds. AI can quickly screen for candidates with a high probability of success and predict efficacy and safety in advance, thereby reducing the time, cost, and risk of failure associated with new drug development.

Through this platform, LG Corp. has integrated and standardized previously dispersed new drug research data, including compound and genomic information, experimental results, academic papers, and patents. Researchers can access the necessary data on a single platform and utilize AI analysis and prediction features with just a few clicks.

The platform provides step-by-step support for the key stages of new drug development. In the stage of identifying the causes of disease, it uses AI to analyze and visualize gene information by cell type and tissue localization, helping researchers identify therapeutic targets.

During the candidate compound design phase, generative AI designs new molecular structures that meet specific criteria. In the validation phase, simulations are used to predict factors such as binding potential between candidate compounds and targets, as well as stability of action, to screen for promising candidates. This approach involves conducting virtual validation in a computer environment prior to actual experiments.

Continuous performance improvement of the AI model is also possible. By comparing AI prediction results with actual experimental data and incorporating these findings into model retraining, prediction performance can be refined as research data accumulates.

LG Corp. has integrated the entire process—from data collection to AI analysis, experimentation, and validation—to accelerate the speed and improve the quality of new drug development, while enabling the utilization of accumulated research data as an asset. Considering the sensitive nature of new drug research data, the company has also established a management system capable of addressing the regulatory and security requirements of the pharmaceutical industry.

This achievement was made possible by LG CNS’s accumulated expertise in AI Transformation (AX) for the pharmaceutical and biotech sectors. LG CNS has been actively pursuing pharmaceutical and biotech AX initiatives, including participation in the Ministry of Health and Welfare’s “K-AI New Drug Development Preclinical and Clinical Model Development Project” and the development of CHONGKUNDANG’s “Agentic AI-Based Annual Quality Assessment Report Preparation Service.”

Recently, the company has been expanding its related business by leveraging “AgenticWorks for Bio,” an Agentic AI platform specialized for the pharmaceutical and biotech sectors.

Lee Jae-seung, Executive Vice President in charge of the Cloud Business at LG Corp., stated, “Through differentiated AX technologies, including Agentic AI, and our specialized expertise in the pharmaceutical and biotech sectors, we will contribute to innovation in AI-driven new drug research and development within the domestic pharmaceutical and biotech industry.” He added, “We will actively support our customers in strengthening their competitiveness in new drug development.”