XtalPi Debuts Kodexia™, the World's First Closed-Loop siRNA Platform Combining Generative AI with First-Principles Modeling

• This unified platform integrates sequence design, chemical modification, experimental validation, extrahepatic delivery design, and intellectual property strategy into an integrated computational–experimental workflow, pioneering in vivo efficacy prediction models to accelerate drug development.
• Validating this infrastructure, XtalPi announced the advancement of six proprietary preclinical programs targeting complex diseases, with the first one reaching NHP data in just 7 months; in utilizing optimized ligand-based delivery mechanisms, XtalPi is advancing RNA interference therapies beyond the liver.
• Internal preclinical data benchmarking demonstrates a nearly threefold increase in molecular design efficiency compared to conventional workflows, alongside a first-round in vivo hit rate exceeding 50% across multiple target programs.

BOSTON and BEIJING, Oct. 6, 2026 /PRNewswire/ -- XtalPi Holdings Limited (HKEX: 2228), an AI and robotics-driven drug and materials discovery company, today debuted Kodexia™, the industry's first closed-loop small interfering RNA (siRNA) discovery platform combining generative AI, first-principles-driven biological modeling, and automated experimentation. Validating the rapid execution capabilities of this new infrastructure, the company simultaneously announced the advancement of six proprietary therapeutic programs, marking the successful expansion of XtalPi's technology into nucleic acid therapeutics.

By silencing disease-driving genes upstream of protein production, siRNA therapeutics offer immense potential for biological targets that conventional modalities struggle to address. However, clinical development remains heavily constrained by fragmented sequence and modification design processes, restrictive patent barriers, inconsistent translation from cellular assays to animal models, and severe limitations in extrahepatic delivery.

To address these interdependent bottlenecks, XtalPi drew on domain-specific expertise to develop siFormer, the proprietary architecture at the core of Kodexia™. By seamlessly integrating RNAi biological mechanisms with nucleic acid chemistry principles, siFormer delivers a qualitative leap in predictive accuracy that surpasses conventional design baselines. The architecture incorporates RNA thermodynamics and structural features relevant to strand loading, target accessibility, and silencing efficiency. These mechanistic constraints guide molecular generation toward biologically plausible designs rather than relying on sequence correlations alone. Through this unified approach, Kodexia™ enables researchers to systematically balance potency, in vivo translation, durability, off-target activity, safety, and patentability instead of optimizing single endpoints in isolation.

To execute this multiparameter optimization, XtalPi utilizes a closed-loop process of computational design, automated experimentation, and empirical feedback. Generative models and the proprietary siFormer architecture operate continuously with XtalPi's automated laboratories, which conduct more than 500 in vitro and 30 in vivo experiments weekly. This seamless integration enables the parallel development of multiple target programs, yielding a nearly threefold increase in design efficiency. Notably, more than 50% of first-round designs tested across multiple programs showed stronger in vivo activity than positive controls.

Kodexia™ also globally co-optimizes dual-target siRNAs as single molecules, creating a strategy for disease biology that may not be adequately addressed through single-target intervention. For extrahepatic delivery, the platform co-designs siRNA molecules with their targeting systems and applies multi-objective optimization to antibody, peptide, and small-molecule conjugates. These capabilities are being applied to delivery research targeting the kidney, spleen, and adipose tissue.

The platform's lead program in immunoglobulin A (IgA) nephropathy serves as an early empirical validation of this unified architecture. The program reached non-human primate efficacy data within seven months, demonstrating greater activity and durability than a clinical-stage reference molecule targeting the same target. The program is on track for preclinical candidate selection within a nine-month project timeline, significantly accelerating the industry norm of 12 to 18 months.

Kodexia™ currently drives six proprietary programs across metabolic, renal, respiratory, and central nervous system diseases, with more than half having already completed in vivo efficacy evaluations. Each internal program generates high-value experimental data that feeds directly back into the platform's models, reinforcing a continuous cycle of discovery and commercial translation. As a global scientific utility platform, XtalPi leverages this capability alongside its established expertise in small molecules, antibodies, and peptides. By advancing these internal assets in tandem with established industry collaborations, XtalPi empowers its partners to pursue high-value targets using the modality best suited to the mechanistic and biophysical principles of RNA interference, including RNA thermodynamics and structural features, ultimately accelerating the delivery of next-generation RNA therapies for patients worldwide.

About XtalPi

XtalPi Holdings Limited (XtalPi, 2228.HK) was founded in 2015 by three physicists from the Massachusetts Institute of Technology (MIT). It is an innovative R&D platform powered by quantum physics, artificial intelligence, and robotics. By integrating first-principles calculations, AI algorithms, high-performance cloud computing, and standardized automation systems, XtalPi provides digital and intelligent R&D solutions for companies in the pharmaceutical, materials science, agricultural technology, energy, new chemicals, and cosmetics industries.

SOURCE XtalPi Inc.