On September 1, 2026, Shimadzu Corporation launched “Methodintelligence for LC,” a cloud service powered by AI agents that analyze scientific literature to propose initial liquid chromatography analysis conditions. The service addresses the significant burden of literature research and condition evaluation in pharmaceutical development, driving greater efficiency in R&D.

Service Launch Overview and Key Features

On September 1, 2026, Shimadzu Corporation released “Methodintelligence for LC” in Japan, a cloud service that utilizes AI to recommend initial analytical conditions for liquid chromatography (LC). The annual base license fee starts at 2.2 million yen (including tax). By simply entering basic inputs such as compound names and analysis objectives, an autonomous AI agent searches and analyzes relevant scientific literature to automatically generate and suggest optimal initial analytical conditions. The underlying database comprises over 40 million highly reliable reference sources, including PubMed, the Japanese Pharmacopoeia, and Shimadzu’s proprietary “Application News.” Shimadzu developed a prototype in 2025, refining the system through visitor trial runs at its Healthcare R&D Center and feedback from industry exhibitions prior to its commercial release. The software was also featured at “JASIS 2026” (September 2–4, 2026, Makuhari Messe), the largest analytical instrument exhibition in Asia. The following screen image demonstrates a typical operational flow. Even early-career researchers can swiftly obtain appropriate conditions, facilitating the standardization of analytical operations across the entire laboratory.

Background and Challenges in Traditional Liquid Chromatography (LC) Analysis

In R&D environments across pharmaceutical and chemical sectors, liquid chromatography analysis is an indispensable process for quantitative and qualitative evaluation of novel compounds. Establishing optimal analysis conditions requires evaluating numerous parameters, including reagent selection, gradient settings, and instrument configurations. Traditionally, experienced researchers had to manually collect, compare, and synthesize insights from vast volumes of published papers, technical reports, and internal archives—a labor-intensive process requiring extensive domain expertise. Moreover, research institutions and enterprises face growing shortages of specialized talent alongside heightened demands for operational efficiency. Consequently, there is an urgent need to establish environments where researchers can focus on creative, high-value R&D rather than routine information gathering. By automating and enhancing literature research, this AI solution addresses the critical need to transition away from knowledge structures dependent on individual experience and intuition, boosting overall research productivity.

How the AI Agent Analyzes Over 40 Million Documents: Mechanisms and Capabilities

Rapid Condition Recommendations Grounded in Trusted Data Sources

At the heart of the service is an AI agent capable of extracting and organizing required analysis conditions from vast literature repositories. Referenced data sources include the global medical literature database PubMed (accessed via Europe PMC), official standards from the Japanese Pharmacopoeia, and Shimadzu’s extensive “Application News” archive accumulated over many years. When a user inputs a target compound name, the AI does not merely output numerical parameters; it presents the underlying reasoning process alongside the supporting literature and compound data directly on screen. This transparency allows researchers to evaluate the AI’s recommendations based on rigorous scientific evidence. Additionally, the system features a seamless CSV export capability for compiled literature lists and proposed conditions, enabling direct integration into internal reports and experimental plans, thereby significantly reducing documentation time. The diagram below illustrates how data is exported and integrated. In this way, the system combines intuitive operation with complete algorithmic transparency.

Visualizing Decision Processes and Fostering Organizational Knowledge Transfer

Method development in chromatography often suffers from reliance on implicit knowledge and individual researcher expertise. Methodintelligence for LC resolves this by securely logging initial conditions generated by the AI alongside user modification histories on the cloud, enabling centralized sharing and management across the organization. This transforms tacit, individual know-how into a shared intellectual asset, facilitating smooth technology transfer and accelerating the onboarding of junior staff. Furthermore, robust measures address information security concerns prevalent in pharmaceutical companies and research institutions. Compound information and search queries entered into the system are strictly excluded from AI retraining datasets. This architecture eliminates the risk of exposing unpublished novel compounds or proprietary research data, ensuring secure operation within strict confidentiality frameworks.

Practical Significance and Strategic Vision for Pharma and Chemistry

Shimadzu’s AI Strategy and Track Record

Shimadzu has a strong track record of deploying software solutions that integrate analytical instrumentation with advanced AI technologies. Notable examples include “LabSolutions MD” for method optimization, as well as “Peakintelligence for LC,” which incorporates AI algorithms trained on extensive analytical data from Ajinomoto Co., Inc. Peakintelligence automates peak integration with precision comparable to expert analysts (approximately 90% matching rate), reducing processing time by roughly 75%. While existing solutions primarily target the measurement and data processing stages—such as peak integration and parameter optimization—the newly released “Methodintelligence for LC” introduces a major innovation by assisting the pre-experimental phase of literature search and initial method drafting. Through these continuous innovations, Shimadzu is actively building a total solution that supports the complete analytical workflow from preparation through data analysis.

Laboratory Workstyle Reform and Future Directions

Laboratory Digital Transformation (Lab DX) has become an imperative in modern research facilities. By automating manual literature searches, Methodintelligence for LC dramatically reduces research time and trial-and-error cycles, accelerating R&D in drug discovery and advanced materials. Freeing scientists from repetitive information gathering allows them to redirect effort toward strategic tasks such as deep analysis and drug target identification, driving meaningful workstyle reform. Looking ahead, industry attention is focused on the evolution of autonomous AI agents interfacing directly with analytical instruments and laboratory execution systems—promising a seamless future where initial condition setup, automated measurement, and data analysis are entirely integrated.

Reference Pages

• [Shimadzu Product Information: “Methodintelligence for LC”] https://www.an.shimadzu.co.jp/products/software-informatics/software-option/methodintelligence-for-lc/index.html
• [Shimadzu Press Release (Prototype Announcement)] https://www.an.shimadzu.co.jp/news-events/campaign/2025/methodintelligence-for-lc.html

【#Shimadzu #GenerativeAI #AIAgent #LiquidChromatography #ResearchAndDevelopment #CloudService #LabDX】