XTALPI and Pfizer Deepen Strategic Collaboration to Co-Develop an AI-Powered Molecular Simulation Platform for Drug and Advanced Materials Discovery

Today, XtalPi (Stock Code:2228.HK) announced the expansion of its strategic partnership with global biopharmaceutical giant Pfizer to jointly develop a next-generation molecular simulation platform aimed at driving drug discovery. This collaboration aims to further integrate the precision of advanced physical algorithms with the speed and scalability of cutting-edge AI models, thereby comprehensively enhancing the efficiency, accuracy, and breadth of chemical space coverage of the molecular simulation platform, significantly accelerating the discovery and development process of small-molecule drugs. Building on the existing collaborative R&D achievements from XtalPi and Pfizer's prior strategic partnership, the two companies will enhance and upgrade the current platform to create a system that combines high precision with higher throughput advantages.Measurement tools to address Pfizer's rapidly expanding small-molecule chemical space and continuously upgrading R&D needs.

In an academic paper jointly published in 2024In the paper¹XtalPi and Pfizer demonstrated the technical advantages of their jointly developed and validated high-precision force field (XtalPi Force Field, XFF). Compared to the most widely used commercial software currently on the market, XtalPi's molecular simulation platform exhibits superior performance across multiple research and development scenarios. This includes accurately predicting small molecule geometries based on molecular mechanics calculations and achieving high-precision affinity predictions through Free Energy Perturbation (FEP) calculations. These core capabilities are crucial for precisely evaluating key drug-like properties of candidate drugs, such as activity, selectivity, and drug resistance. Additionally, the molecular simulation platform can cover a larger chemical space, expanding the range of molecules that can be modeled with precision.The Scope of Screening and Rational Design;Higher computational throughput brings significant cost reduction, enabling large-scale evaluation of candidate structures with high-precision methods at an earlier stage of R&D and expanding a wider range of R&D application scenarios.

In the statistical analysis of predictive performance across over 30 datasets and more than 1000 molecular pairs, compared with the commercial software with the largest global market share (corresponding to the dark blue "FEP+" results in the chart), XtalPi'sXFEP andThe XFF combination (corresponding to the light blue "XFF" result in the chart) performs better in high-precision affinity prediction.（Xue et al., 2024）。

After the upgrade of this cooperation, the teams from both sides will develop more accurate predictive models for Pfizer's proprietary chemical space and expand the platform’s broader applications in the field of small molecule drug R&D. XtalPi will locally deploy its self-developed XFEP for Pfizer.Free Energy Perturbation Platform, covering the entire process from parameter customization to FEP calculations, fully empowering Pfizer's drug discovery efforts. The platform aims to achieveTargeting higher precision, higher throughput, and better user experience, it will provide Pfizer scientists with cross-scenario drug design and development.IssueSupport.

"HuiPfizer's leadership and profound scientific research accumulation in the field of pharmaceutical innovation have injected strong momentum into the continuous evolution of the XtalPi AI platform.” XtalPi CEO Dr. Ma Jian said。"ThisThe collaboration once again confirms that XtalPi's R&D platform, which integrates physics insights with cutting-edge AI technology, represents a highly promising technical development direction, also driving us to continuously develop more powerful predictive tools.Bring profound changes to the industryThrough in-depth cooperation with the Pfizer team, we will jointly build a new generation of drug research and development platform, break through the limitations of traditional R&D models, bring more potential and challenging R&D tasks within range, and accelerate the delivery of breakthrough therapies to patients worldwide.” 

¹ Xue, B., Yang, Q., Zhang, Q., et al. (2024). Development and comprehensive benchmark of a high-quality AMBER-consistent small molecule force field with broad chemical space coverage for molecular modeling and free energy calculation. Journal of Chemical Theory and Computation, 20(2), 799–818. https://doi.org/10.1021/acs.jctc.3c00920