Advancing drug discovery from automation to autonomy: Insilico Medicine announces LabClaw The Intelligent System

Insilico Medicine ( “Insilico”, HKEX:3696 ), a clinical-stage, generative AI–driven drug discovery company, today announced LabClaw, its next-generation intelligent laboratory operating system. As the pharmaceutical industry's first laboratory autonomy system built on a lightweight Agent-Guard architecture, LabClaw is specifically designed to complement Insilico’s fully automated laboratory, LifeStar2, aiming to drive a paradigm shift in intelligent laboratories from instruction execution to autonomous coordination.

 

The system establishes a novel three-pillar experimental paradigm consisting of AI thinking, automated execution, and human supervision, integrating PandaOmics, Insilico's proprietary AI target discovery platform, automated laboratory hardware matrix, Laboratory Information Management System (LIMS), and multi-dimensional data analysis pipelines. With 5 collaborative AI agents and 28 specialized skill modules, LabClaw established an end-to-end intelligent closed loop spanning therapeutic target discovery, compound screening, automated experimental execution, data analysis, and report generation across computational and wet-lab experiments. Meanwhile, the system incorporates a Human-in-the-Loop approval mechanism at critical junctures, ensuring scientific and compliance requirements are met, while enhancing intelligent autonomy.

“The core significance of LabClaw lies in reshaping scientists' focus. By taking over time-consuming standardized operations and tedious scheduling coordination, LabClaw serves as a tireless research co-pilot”, says Lin Sha, Senior Director of IT at Insilico Medicine. "Here, the Human-in-the-Loop mechanism ensures scientists remain at the center of decision-making. It frees brilliant minds from repetitive pipetting, allowing them to devote their most valuable energy to generating scientific hypotheses, extracting insights from complex data, and designing innovative strategies."

Breaking Process Rigidity: From Fragmented Automation to Intelligent Autonomy

Over the past decade, laboratory automation in biomedicine has advanced significantly. The adoption of liquid handling workstations, high-throughput screening platforms, automated cell culture systems and other precision equipment has dramatically increased experimental throughput, and effectively reduced systematic errors compared to traditional manual operations. However, the industry still faces three primary bottlenecks in moving from automation to true autonomy:

• Workflow Rigidity: Pre-programmed workflows often struggle with dynamic variables. If an intermediate step fails or a QC anomaly occurs, the system would demand manual intervention.

• Data Silos: Without automatic interpretation and closed-loop feedback pathways, the massive data output from equipment lacks connection to the upstream AI-generated inferences and downstream statistical analysis, holding real-time data usage and value realization.

• High Coordination Costs: Multi-step experiments across different functional islands still require manual orchestration and tracking, creating a productivity bottleneck with substantial non-research tasks.

 

To address these challenges, Insilico set out to build an AI agent matrix capable of deeply understanding the scientific context, autonomously coordinating cross-device processes, and collaborating efficiently with human scientists at critical decision points. LabClaw is the engineering realization of this vision, not only integrating the advanced hardware equipment of Insilico's LifeStar2 laboratory but also establishing an intelligent brain that actually understands research intent.

At the core of LabClaw is the original Agent-Guard architecture. The system features five types of agents assigned to specialized tasks: Experiment Coordinator, Orchestration Expert, Science Analyst, QC Inspector, and Data Specialist, while maintaining a Human-in-the-Loop confirmation mechanism at critical operational checkpoints. This design allows agents freedom to operate within safe boundaries while ensuring all status changes are validated by the backend and confirmed by a human, maximizing efficiency without compromising scientific safety standards.

 

Conversational Laboratory: Starting with a Research Question

LabClaw introduces conversational experimental design to the laboratory. Researchers can initiate tasks using natural language without needing programming expertise, and the built-in AI agents autonomously understand the goal before automatically establishing complete workflows around it, which includes target selection, reagent preparation, equipment execution, and data analysis. Users can also interact with the platform at embedded checkpoints to provide expert insights or fine-tune parameters for AI-empowered virtual screening or experimental protocols.

In execution mode, the system automatically dispatches the PandaOmics engine and LIMS system, coordinating the six major functional islands in the LifeStar2 laboratory, including cell culture, high-throughput screening, and next-generation sequencing across dozens of devices to advance experimental workflows, while AGVs (Automated Guided Vehicles) coordinate inter-island material transfers, achieving intelligent autonomy from analysis to execution of research tasks.

In terms of data integration, LabClaw also demonstrates powerful integration capabilities. It automatically completes CRISPR editing efficiency analysis, RT-qPCR calculations, and NGS quality control, feeding validated data back to the proprietary PandaOmics platform in real-time. This creates a closed loop from computational prediction to wet-lab validation and model optimization, where AI learns from experimental feedback to identify more promising drug targets and molecules with improving precision.

LabClaw encapsulates the entire drug development process into 20 modular, Lego-like steps covering AI/computational services, sample and material preparation, automated experimental execution, data analysis and report generation, and notification collaboration. The system also features 8 pre-configured, standardized experimental plan templates for core scenarios including target validation, CRISPR biomarker validation, high-throughput screening, hit molecule discovery, phenotypic safety assessment, LifeStar2 operations, LIMS material readiness checks, and data analysis, while allowing researchers to customize and save their own private or public templates through dialogue for future reusage.

 

End-to-End Acceleration in Practice: A Case Study of Anti-Aging Target Discovery Closed-Loop Validation

 

In a recent anti-aging target discovery and intelligent screening case, the research team utilized LabClaw’s Science Analyst agent to move beyond the previously inefficient mode of reviewing massive literature and manually organizing evidence. By simply issuing a request via dialogue, the system queried PandaOmics and multi-source knowledge bases to screen the 20 most promising targets, providing key supporting evidence for each target including scores, novelty assessments, and known drug associations. Researchers then confirmed the selection on the interface to finish target identification from massive search, compressing weeks of biological background research into minutes.

 

Next, the project advances to the experimental design and workflow orchestration phase, where LabClaw demonstrates its ability to directly translate scientific concepts into executable experiments. After target identification, the system's Experiment Coordinator agent automatically orchestrates a 7-step standard operating procedure including material readiness checks, cell culture, gene editing, and high-content imaging, achieving automatic coordination and consistent implementation of process linkages. Meanwhile, the QC Inspector agent provides full-process monitoring of environmental parameters and cell states, significantly reducing data fluctuations from experimental operations and improving process stability and precision.

 

In the closed-loop validation and data feedback phase, LabClaw establishes an automated chain from execution to analysis and feedback. After equipment automatically completes plating, transfection, imaging, and other tasks, the system immediately takes over data processing, automatically calculating gene editing efficiency and cell phenotype scores, and transmitting results in real-time to the database for dynamic update of target importance rankings. Throughout the entire process, researchers only need to make decision confirmations at three critical junctures, with all other repetitive, foundational experimental work automatically completed by the system, thereby significantly accelerating research progress and releasing scientists' energy back to more core scientific judgment and strategic design.

 

Conclusion

 

LabClaw extends Insilico Medicine's pharmaceutical superintelligence (PSI) boundary from the digital world to the physical world, bridging the final gap between algorithmic prediction and laboratory validation. By integrating target discovery, molecular generation, and experimental validation, LabClaw forms an end-to-end AI-driven loop from computational analysis to wet lab validation, advancing drug development from fragmented automation into intelligent autonomous cycles.

 

“LabClaw delivers not only enhanced R&D efficiency but represents an upgrade in research paradigm: scientists are liberated from heavy experimental orchestration and process coordination to focus their energy on more creative scientific judgment and research”, says Alex Zhavoronkov PhD, Founder and CEO of Insilico Medicine. “Looking ahead, with further integration of multi-laboratory collaboration, adaptive optimization capabilities, and compliance systems, we hope to build an open, intelligent, and seamlessly connected drug development ecosystem, opening a new chapter in AI-driven drug discovery and advancing toward pharmaceutical superintelligence.”

 

About Insilico Medicine 

 

Insilico Medicine is a pioneering global biotechnology company dedicated to integrating artificial intelligence and automation technologies to accelerate drug discovery, drive innovation in the life sciences, and extend healthy longevity to people on the planet. The company was listed on the Main Board of the Hong Kong Stock Exchange on December 30, 2025, under the stock code 03696.HK. 

 

By integrating AI and automation technologies and deep in-house drug discovery capabilities, Insilico is delivering innovative drug solutions for unmet needs including fibrosis, oncology, immunology, pain, and obesity and metabolic disorders. Additionally, Insilico extends the reach of Pharma.AI across diverse industries, such as advanced materials, agriculture, nutritional products and veterinary medicine. For more information, please visit www.insilico.com

 

---

---