GPX4

BEIJING, China I January 8, 2025 I In a recent achievement, DP Technology’s hit discovery platform RiDYMO® has successfully designed a cyclic peptide targeting β-catenin, a historically “undruggable” protein, in just two months. This innovative AI-driven approach combines advanced algorithms, physics simulations, and high-throughput experimentation to rapidly develop potential drug candidates for challenging targets. The platform’s success with β-catenin, a key player in cancer development, demonstrates its potential to revolutionize drug discovery for previously intractable targets. Fig. 1 RiDYMO® : an AI for Science-Driven Hit Discovery and Optimizaton Platform Fig. 2 The workflow of the Cyclic Peptide Screening Process Based on the RiDYMO® Platform Fig. 3 Experimental Results of 29 Cyclic Peptide Molecules The Wnt/β-catenin pathway plays a crucial role in various biological functions and is implicated in multiple cancers[1-3]. Despite its importance as a therapeutic target[4-7], β-catenin has long been considered “undruggable” due to its flat protein surface[8-10]. The RiDYMO® platform overcomes this challenge by leveraging a vast library of over 10^12 cyclic peptides and macrocyclic molecules, incorporating both natural and non-natural amino acids. Using the RiDYMO® platform, researchers rapidly designed and screened cyclic peptide molecules for β-catenin binding. The process involved: The results were impressive, with 12 compounds showing strong inhibition of protein-protein interactions (IC50≤10 μM), a success rate of over 40%. One standout compound, designated as compound 24, effectively bound to β-catenin, disrupted its interaction with BCL9, and inhibited the Wnt/β-catenin signaling pathway. Dr. Dongdong Wang, Co-President of DP Technology’s Drug Discovery Unit, emphasized the platform’s ability to tackle undruggable targets and rapidly design high-quality compounds. The RiDYMO® platform’s success with β-catenin demonstrates its potential to accelerate the development of orally available cyclic peptide drugs, potentially transforming the pharmaceutical industry’s approach to challenging targets. This breakthrough in cyclic peptide design for β-catenin showcases the power of AI-driven drug discovery platforms in addressing previously intractable targets, potentially opening new avenues for cancer treatment and other therapeutic areas. About the RiDYMO® Hit Discovery and Optimization Platform RiDYMO® is a state-of-the-art hit discovery and optimization platform developed by DP Technology, leveraging the principles of AI for Science. It employs the proprietary Hermite® computational drug design software to elucidate the dynamics of “undruggable” targets and explores a broader chemical space encompassing small molecules, macrocycles, and cyclic peptides. By integrating advanced artificial intelligence, physics-based algorithms, and high-throughput experimentation, the platform excels in designing oral macrocyclic compounds and rapidly delivers innovative drug candidates. As one of its core algorithms, Reinforced Dynamics (RiD)[4] has a significant advantage in the sampling efficiency of molecular dynamics simulation. By fully leveraging the high-dimensional representation capabilities of neural networks, RiD can efficiently capture dynamic conformational changes in complicated biomolecular systems. Previously, the core RiD algorithm of the platform was published in Nature Computational Science [11]. The study demonstrated that RiD could achieve a more comprehensive free energy surface within 1.86 μs, compared to 100 μs required by traditional MD methods, representing nearly a hundredfold increase in efficiency. RiDYMO® has been successfully employed in various drug discovery programs, including the development of small molecule compounds for crucial targets such as c-Myc and GPX4, as well as other modalities such as cyclic peptides and ADCs. For more information, please visit our website. ( https://www.dp.tech/en/services/medicine ). About Hermite Hermite® is a next-generation computational drug design platform developed by DP Technology, powered by AI for Science, to provide a comprehensive solution for drug design. Hermite® integrates industry-leading tools such as the Free Energy Perturbation module (“Uni-FEP”), and the ultra-high-throughput virtual screening tool (“Uni-VSW”). The platform supports key stages of drug discovery from protein structure prediction and target validation to hit discovery and lead optimization. In addition, the platform offers an interactive, web-based molecular visualization experience, with detailed management of projects, teams, and data. It features full compliance certification, multi-tier security measures. Users have the flexibility to choose between cloud-based or private deployment options. Hermite® is rapidly gaining traction in the pharmaceutical industry, with over 60% of leading companies in China utilizing the platform across more than 50 drug pipelines. This growing trust is reflected in the impressive number of calculations performed—over 200,000 Uni-FEP calculations to date—demonstrating Hermite®’s reliability and effectiveness as a powerful tool for drug discovery and development. For more information, please visit: https://www.dp.tech/en/product/hermite References: SOURCE: DP Technology

BEIJING, Jan. 8, 2025 /PRNewswire/ -- In a recent achievement, DP Technology's hit discovery platform RiDYMO® has successfully designed a cyclic peptide targeting β-catenin, a historically "undruggable" protein, in just two months. This innovative AI-driven approach combines advanced algorithms, physics simulations, and high-throughput experimentation to rapidly develop potential drug candidates for challenging targets. The platform's success with β-catenin, a key player in cancer development, demonstrates its potential to revolutionize drug discovery for previously intractable targets. Continue Reading Fig. 1 RiDYMO® : an AI for Science-Driven Hit Discovery and Optimizaton Platform Fig. 2 The workflow of the Cyclic Peptide Screening Process Based on the RiDYMO® Platform Fig. 3 Experimental Results of 29 Cyclic Peptide Molecules The Wnt/β-catenin pathway plays a crucial role in various biological functions and is implicated in multiple cancers[1-3]. Despite its importance as a therapeutic target[4-7], β-catenin has long been considered "undruggable" due to its flat protein surface[8-10]. The RiDYMO® platform overcomes this challenge by leveraging a vast library of over 10^12 cyclic peptides and macrocyclic molecules, incorporating both natural and non-natural amino acids. Using the RiDYMO® platform, researchers rapidly designed and screened cyclic peptide molecules for β-catenin binding. The process involved: Optimizing the β-catenin protein structure to improve the binding interface. High-throughput screening and design of cyclic peptides using the platform's extensive library. Synthesis and testing of 29 candidate molecules. The results were impressive, with 12 compounds showing strong inhibition of protein-protein interactions (IC50≤10 μM), a success rate of over 40%. One standout compound, designated as compound 24, effectively bound to β-catenin, disrupted its interaction with BCL9, and inhibited the Wnt/β-catenin signaling pathway. Dr. Dongdong Wang, Co-President of DP Technology's Drug Discovery Unit, emphasized the platform's ability to tackle undruggable targets and rapidly design high-quality compounds. The RiDYMO® platform's success with β-catenin demonstrates its potential to accelerate the development of orally available cyclic peptide drugs, potentially transforming the pharmaceutical industry's approach to challenging targets. This breakthrough in cyclic peptide design for β-catenin showcases the power of AI-driven drug discovery platforms in addressing previously intractable targets, potentially opening new avenues for cancer treatment and other therapeutic areas. About the RiDYMO® Hit Discovery and Optimization Platform RiDYMO® is a state-of-the-art hit discovery and optimization platform developed by DP Technology, leveraging the principles of AI for Science. It employs the proprietary Hermite® computational drug design software to elucidate the dynamics of "undruggable" targets and explores a broader chemical space encompassing small molecules, macrocycles, and cyclic peptides. By integrating advanced artificial intelligence, physics-based algorithms, and high-throughput experimentation, the platform excels in designing oral macrocyclic compounds and rapidly delivers innovative drug candidates. As one of its core algorithms, Reinforced Dynamics (RiD)[4] has a significant advantage in the sampling efficiency of molecular dynamics simulation. By fully leveraging the high-dimensional representation capabilities of neural networks, RiD can efficiently capture dynamic conformational changes in complicated biomolecular systems. Previously, the core RiD algorithm of the platform was published in Nature Computational Science [11]. The study demonstrated that RiD could achieve a more comprehensive free energy surface within 1.86 μs, compared to 100 μs required by traditional MD methods, representing nearly a hundredfold increase in efficiency. RiDYMO® has been successfully employed in various drug discovery programs, including the development of small molecule compounds for crucial targets such as c-Myc and GPX4, as well as other modalities such as cyclic peptides and ADCs. For more information, please visit our website. (). About Hermite Hermite® is a next-generation computational drug design platform developed by DP Technology, powered by AI for Science, to provide a comprehensive solution for drug design. Hermite® integrates industry-leading tools such as the Free Energy Perturbation module ("Uni-FEP"), and the ultra-high-throughput virtual screening tool ("Uni-VSW"). The platform supports key stages of drug discovery from protein structure prediction and target validation to hit discovery and lead optimization. In addition, the platform offers an interactive, web-based molecular visualization experience, with detailed management of projects, teams, and data. It features full compliance certification, multi-tier security measures. Users have the flexibility to choose between cloud-based or private deployment options. Hermite® is rapidly gaining traction in the pharmaceutical industry, with over 60% of leading companies in China utilizing the platform across more than 50 drug pipelines. This growing trust is reflected in the impressive number of calculations performed—over 200,000 Uni-FEP calculations to date—demonstrating Hermite®'s reliability and effectiveness as a powerful tool for drug discovery and development. For more information, please visit: References: Salik B, et al. Targeting RSPO3-LGR4 signaling for leukemia stem cell eradication in acute myeloid leukemia. Cancer Cell. 2020;38(2):263–78.e6. Liu J, et al. Wnt/β-catenin signal ling: function, biological mechanisms, and therapeutic opportunities. Signal Transduct Target Ther. 2022 Jan 3;7(1):3. Choi BR, Cave C, Na CH, Sockanathan S. GDE2-dependent activation of canonical wnt signaling in neurons regulates oligodendrocyte maturation. Cell Rep. 2020;31(5):107540. Zhang L, Shay JW. Multiple roles of APC and its therapeutic implications in colorectal cancer. J. Natl. Cancer Inst. 2017;109(8). Belenguer G, et al. RNF43/ZNRF3 loss predisposes to hepatocellular-carcinoma by impair ing liver regeneration and altering the liver lipid metabolic ground-state. Nat. Commun. 2022;13(1):334. Cao L, et al. Helicobacter pylori-induced RASAL2 Through Activation of Nuclear Factor-κB Promotes Gastric Tumorigenesis via β-catenin Signaling Axis. Gastroenterology. 2022 May;162(6):1716-1731.e17. Hashemi M, et al., Biological functions and molecular interactions of Wnt/β-catenin in breast cancer: Revisiting signaling networks. Int. J. Biol. Macromol. 2023 Mar 31:232:123377. Hwang et al., Direct Targeting of b-Catenin by a Small Molecule Stimulates Proteasomal Degradation and suppresses Oncogenic Wnt/β-Catenin Signaling. 2016, Cell Rep. 16, 28–36. Tanton et al., A novel β-catenin/BCL9 complex inhibitor blocks oncogenic Wnt signaling and disrupts cholesterol homeostasis in colorectal cancer. 2022, Sci. Adv. 8, eabm3108. Scuoppo C. et al., "ST316, a Clinical Peptide Antagonist of β-catenin, Induces Anti-Tumor Immune Responses by Multiple Mechanisms of Action." AACR 2024. Wang D, et al. Efficient sampling of high-dimensional free energy landscapes using adaptive reinforced dynamics. 2022, Nat. Comp. Sci. 2, 20. SOURCE DP Technology WANT YOUR COMPANY'S NEWS FEATURED ON PRNEWSWIRE.COM? 440k+ Newsrooms & Influencers 9k+ Digital Media Outlets 270k+ Journalists Opted In GET STARTED

BEIJING, Nov. 19, 2024 /PRNewswire/ -- With the support of the RiDYMO® platform, we are able to comprehensively explore various potential targets in drug development, including protein-protein interactions, kinases, GPCRs, ion channels, and transcription factors. Continue Reading Figure: Screening process for novel scaffold lead compound Figure A: Optimization and FEP prediction for R1, R2, and R3, with dark gray areas indicating a 1 kcal error range and light gray areas representing a 2 kcal error range; B: Overall molecular optimization process based on the DP101 scaffold Figure: Antitumor effects of DP226 on HCT116 xenograft model Polo-like kinase 1 (PLK1) is a highly conserved serine/threonine protein kinase that plays a crucial role during cellular mitosis[1]. It is overexpressed in various malignancies, including colorectal cancer[2], pancreatic cancer[3], and breast cancer, and its high expression is associated with poor prognosis. Consequently, PLK1 has emerged as a promising target in cancer drug development, with numerous candidates currently in the pipeline worldwide. Among these, Cardiff Oncology's onvansertib (PCM-075) has shown the most significant progress, displaying encouraging efficacy results in clinical treatment for patients with metastatic colorectal cancer (mCRC). The RiDYMO® platform, a next-generation computational drug design platform developed by DP Technology, based on AI for Science, provides robust end-to-end support for hit discovery and optimization by integrating multiple artificial intelligence techniques, physical algorithms, and experimental validation. In this study, we referred to the co-crystal complex of PLK1 and PCM075 (2YAC) and utilized the RiDYMO platform's comprehensive capabilities to synthesize and test under 100 compounds, leading to the identification of preclinical candidate compounds with potential clinical value and significantly accelerating the drug development process. We first conducted an in-depth analysis of the interactions between PCM-075 and the PLK1 protein. Preserving key interactions, we applied molecular generation, evaluation, and screening techniques to obtain a novel scaffold, the lead compound DP101, which exhibited an activity of 56 nM. Next, during the optimization process based on the structure of DP101, we modified several sites on the scaffold, including hydrogen bonding networks (R3) and hydrophobic group encapsulation zones (R1 & R2). Experimental data showed a high consistency between computational predictions and experimental results, demonstrating the platform's practicality in optimizing kinase inhibitors. Leveraging the RiDYMO® platform's activity prediction ranking and physicochemical property forecasting, we underwent multiple rounds of testing and optimization, ultimately obtaining several compounds with enhanced in vitro activity and pharmacokinetic properties. Notably, the activity of DP226 (0.19 nM) improved nearly 300-fold compared to DP101, surpassing PCM075, with an oral bioavailability (F = 76.8%) significantly higher than that of PCM075 (F = 24%). At low doses, DP226 exhibited superior anti-tumor effects compared to PCM075, and in combination with bevacizumab, it led to tumor regression, positioning it as a potential PCC molecule for further studies. Dr. Dongdong Li, Director of Medicinal Chemistry at DP Technology and Project Leader, stated, "With the support of the RiDYMO® platform, we are able to comprehensively explore various potential targets in drug development, including protein-protein interactions, kinases, GPCRs, ion channels, and transcription factors. The use of the RiDYMO® platform for activity predictions, rankings, and experimental validations significantly enhances the efficiency of activity optimization, effectively shortening optimization cycles and accelerating the overall project development timeline. The integrated capabilities of the RiDYMO® platform hold promise for providing high-efficiency support in early-stage drug discovery. As a potential PCC molecule, we look forward to collaborating with experienced pharmaceutical companies in this field to advance this project to the next milestone." About the RiDYMO® Hit Discovery and Optimization Platform RiDYMO® is a state-of-the-art hit discovery and optimization platform developed by DP Technology, leveraging the principles of AI for Science. It employs the proprietary Hermite® computational drug design software to elucidate the dynamics of "undruggable" targets and explores a broader chemical space encompassing small molecules, macrocycles, and cyclic peptides. By integrating advanced artificial intelligence, physical algorithms, and high-throughput experimentation, the platform excels in designing oral macrocyclic compounds and rapidly delivers innovative drug candidates. As one of its core algorithms, Reinforced Dynamics (RiD)[4] has a significant advantage in the sampling efficiency of molecular dynamics simulation. By fully leveraging the high-dimensional representation capabilities of neural networks, RiD can efficiently capture dynamic conformational changes in complicated biomolecular systems. Previously, the core RiD algorithm of the platform was published in Nature Computational Science. The study demonstrated that RiD could achieve a more comprehensive free energy surface within 1.86 μs, compared to 100 μs required by traditional MD methods, representing nearly a hundredfold increase in efficiency. RiDYMO® has been successfully employed in various projects, including the development of c-Myc small molecules, GPX4 small molecules, and β-catenin cyclic peptides, demonstrating its strong potential in innovative drug discovery. For more information, please visit our website. (). About Hermite Hermite® is a next-generation computational drug design platform developed by DP Technology, powered by AI for Science, to provide a comprehensive solution for drug design. Hermite® integrates industry-leading tools such as the Free Energy Perturbation module, Uni-FEP, and the ultra-high-throughput virtual screening tool, Uni-VSW, supporting key stages of drug discovery, from protein structure prediction and target validation, to hit discovery and lead optimization. The platform offers an interactive, web-based molecular visualization experience, with detailed management of projects, teams, and data. It features full compliance certification, multi-tier security measures, and supports both cloud-based and private deployment options. Hermite® is trusted by over 60% of leading pharmaceutical companies in China, applied across more than 50 drug pipelines. To date, industry users have performed over 200,000 Uni-FEP calculations on the Hermite® platform. For more information, please visit: About DP Technology At DP Technology, we're at the forefront of integrating artificial intelligence into scientific research and industrial R&D. Our "AI for Science" initiative is redefining how we tackle complex scientific challenges, making groundbreaking discoveries more accessible and actionable. We've developed the "DP Particle Universe," a suite of advanced pre-trained models that seamlessly connect cutting-edge research with real-world industrial applications. Our software suite includes: Bohrium® Scientific Computing Space Station, Hermite® Computational Drug Design Platform, RiDYMO® Hit Discovery Platform, and Piloteye® Battery Design Automation Platform. Together, these platforms form a robust foundation for industrial innovation and an open ecosystem for AI in science, fostering advancements in key areas such as drug discovery, energy, materials science and information technology. SOURCE DP Technology WANT YOUR COMPANY'S NEWS FEATURED ON PRNEWSWIRE.COM? 440k+ Newsrooms & Influencers 9k+ Digital Media Outlets 270k+ Journalists Opted In GET STARTED