Alivexis, Inc.

NEWS - 15 July, 2025 - Alivexis, Inc. (Headquartered in Minato-ku, Tokyo; CEO S. Roy Kimura, “Alivexis”) a preclinical-stage computation-driven drug discovery firm, is pleased to announce that it has been selected for the third round of the Ministry of Economy, Trade and Industry (METI) and the New Energy and Industrial Technology Development Organization (NEDO)’s Generative AI Accelerator Challenge (GENIAC) project, which aims to strengthen domestic development capabilities in generative AI. The GENIAC project primarily provides high-performance computational resources to support the development of foundational models, which are core technologies in generative AI, and provides support for proof-of-concept investigations into the practical application of data and AI. The first round of the project was implemented starting in February 2024, and the second round began in October of the same year. In this third round, Alivexis aims to utilize the computational resources provided to develop a generative AI foundational model for use in drug discovery that can predict the biological activity of small molecule compounds with world-leading accuracy by integrating our proprietary simulation technology ModBind™ into the process of AI model training. Through the implementation of the AI model supported by this GENIAC project, we hope to further contribute to the development of innovative medicines and the improvement of drug discovery research productivity. Background In the field of small molecule drug discovery, physics-based simulations for biological activity prediction have been widely applied and have delivered many successful outcomes. While AI is expected to play an increasingly important role, its use in this area faces several challenges, including the vast structural diversity of small molecule compounds and the limitations in the quality of datasets available for AI training, particularly those found in public domain. As a result, current AI models are not yet capable of delivering sufficiently high performance in this field. In contrast, AI has made significant advances in other fields, such as autonomous driving, by leveraging not only real-world data but also synthetic data generated through simulation. This has enabled practical applications with high performance. Drawing on this success, it is expected that simulation-generated data can similarly accelerate progress in AI-driven drug discovery. However, traditional bioactivity prediction methods rely heavily on compounds with known activity to achieve high accuracy. These methods have struggled to significantly improve prediction speed while maintaining precision. To date, no simulation technology has successfully combined the quality, volume, and wide range of adaptability required for effective AI training in this field. In response to these challenges, our proprietary simulation technology, ModBind™, enables high-speed, high-accuracy prediction of absolute biological activity without the need for compounds with known experimental data. This technology has already led to the identification of five clinical candidate compounds. Based on these results, we believe that integrating ModBind™ with AI holds strong potential as an innovative solution to the current limitations of AI-driven small molecule drug discovery. As a result, we have launched the GENIAC initiative to demonstrate this potential in practice. Key Features of the Small Molecule Drug Discovery Foundational AI Model to Be Developed in This Project In this project, we will use ModBind™, our independently developed physics-based simulation algorithm for predicting the biological activity of small molecules, as an “oracle” in an active learning workflow. This approach will allow us to generate highly accurate biological activity prediction data for new compounds, which will serve as additional training data for the foundational AI model. By incorporating active learning techniques into this training process, we will efficiently expand the learning scope across a wide range of compound spaces. This is expected to enhance the adaptability of the model and significantly improve its predictive accuracy. Through this initiative, we aim to develop a foundational AI model for small molecule drug discovery that achieves unprecedented prediction accuracy across multiple drug targets. About Alivexis’ proprietary drug discovery platform ModBind™ ModBind™ is a proprietary physics-based simulation technology developed by Alivexis that enables high-speed, high-accuracy prediction of the absolute binding strength between small molecule compounds and target proteins, without the need for experimental activity data. By leveraging ModBind™ along with other platform technologies, we have already identified 5 clinical candidate compounds. One of these led to a licensing agreement with a Swiss company last year valued at approximately 42.5 billion yen, demonstrating the practical and commercial utility of our technology. In addition, we have conducted multiple joint research projects with pharmaceutical companies, with ModBind™ serving as the central technology in these collaborations. Reference: NEDO Project Announcement Page (Japanese only) https://www.nedo.go.jp/koubo/CD3_100397.html Notice of Selection for the Ministry of Economy, Trade and Industry’s Domestic Generative AI Development Strengthening Project “GENIAC”(PDF) If you’d like more information about this topic, please email at info@alivexis.com

NEWS - 03 March, 2025 - Alivexis, Inc. (Headquartered in Minato-ku, Tokyo; CEO S. Roy Kimura, “Alivexis”) a preclinical-stage computation-driven drug discovery firm, announces the second closing of its Series D round in the amount of 100M JPY (approximately $0.7M USD). Investor participating in the second closing is Ohara Pharmaceutical Co., Ltd. (Headquartered in Koga City, Shiga Prefecture; CEO Seiji Ohara). The total amount raised in Series D round is 900M JPY (approximately $6.0 USD), and Alivexis has raised a total of 6.76B JPY (approximately $45.0M USD) to date. Alivexis will use the proceeds to further research and development programs utilizing its cutting-edge computational science-based drug discovery platform “ModBindTM”, in addition to accelerating over 10 in-house R&D programs. Alivexis will expand key business strategies in early out-licensing and drug discovery collaboration, with the aim of further increasing its corporate value. Comment from S. Roy Kimura, Ph.D., Co-founder and CEO: “We are excited to announce the second closing of our latest funding round by Ohara Pharmaceutical Co., Ltd. Since our last fund-raise, we have successfully licensed our cathepsin C inhibitor program and are currently working together with our partner toward IND submission. We also published a scientific article on the first version of our ModBindTM simulation platform, based on a unique theoretical approach that enables accurate absolute ligand efficacy predictions at ~2000x speeds relative to existing methods. In the next several months, we expect to sign additional licensing partnerships from our portfolio of immunology and oncology R&D programs, as well as additional ModBindTM-related alliances and collaborations. We look forward to continuing our mission with our shareholders and partners to accelerate discovery of new medicines for patients and their families in need.” About Alivexis, Inc.; Name： Alivexis, Inc. Headquarter: Daiichi Hibiya Building 7F, Shimbashi 1-18-21, Minato-ku, Tokyo 105-0004, Japan Representative directors: S. Roy Kimura, Ph.D., CEO / Kazuki Ohno, Dr. Eng. COO Established: August 8, 2016 URL: https://alivexis.com Business Description: A network-based drug discovery company utilizing cutting-edge technologies. About Ohara Pharmaceutical Co., Ltd.; Name: Ohara Pharmaceutical Co., Ltd. Headquarter: 121-15 Toriino, Koga-cho, Koga City, Shiga Prefecture 520-3403, Japan Representative director: Seiji Ohara, CEO Established: November 5, 1964 URL: https://www.ohara-ch.co.jp/ Business Description: Discovery and manufacture of orphan and generic drugs. Focuses on developing new drugs in core therapeutic areas, child cancer, hematologic cancer, CNS and development, manufacturing and marketing of generic drugs. Aims to provide comprehensive healthcare solutions by innovating in treatment, prevention, diagnosis, and aftercare. Alivexis Raises $0.7M USD Series D in Second Close(PDF) If you’d like more information about this topic, please email at info@alivexis.com

NEWS - 07 January, 2025 - Alivexis, Inc. (Headquartered in Minato-ku, Tokyo; CEO S. Roy Kimura, “Alivexis”) a preclinical-stage computation-driven drug discovery firm, announces the first closing of its Series D round in the amount of 800M JPY (approximately $5.1M USD). Investors participating in the first closing include funds managed by JIC Venture Growth Investments Co., Ltd., DBJ Capital Co., Ltd., and Fast Track Initiative, Inc. Alivexis has raised a total of 6.66B JPY (approximately $42.3M USD) to date. Alivexis will use the proceeds to further research and development programs utilizing its cutting-edge computational science-based drug discovery platform “ModBindTM”, in addition to accelerating over 10 in-house R&D programs. Alivexis will expand key business strategies in early out-licensing and drug discovery collaboration, with the aim of further increasing its corporate value. Comment from S. Roy Kimura, Ph.D., Co-founder and CEO: “We are excited to announce the first closing of our latest funding round led by the JIC Venture Growth Investments Co., Ltd. with follow-on investments from existing shareholders. Since our last fund-raise, we have successfully licensed our cathepsin C inhibitor program and are currently working together with our partner toward IND submission. We also published a scientific article on the first version of our ModBindTM simulation platform, based on a unique theoretical approach that enables accurate absolute ligand efficacy predictions at ~2000x speeds relative to existing methods. In the next several months, we expect to sign additional licensing partnerships from our portfolio of immunology and oncology R&D programs, as well as additional ModBindTM-related alliances and collaborations. We look forward to continuing our mission with our shareholders and partners to accelerate discovery of new medicines for patients and their families in need.” Comment from Hanae Suzuki, JIC Venture Growth Investments Co., Ltd.: “We are thrilled to announce our recent investment in Alivexis and are honored to collaborate with their exceptional team, renowned for its extensive expertise in the pharmaceutical a industries and its globally integrated operations. In this era of rapid innovation in AI and semiconductor technologies, we recognize the significant potential of Alivexis’s cutting-edge computational chemistry-based simulation platform. Our firm is deeply committed to supporting Alivexis in driving its continued growth and innovation. We are confident that this investment will not only accelerate the development of groundbreaking medicines for patients and their families but also play a pivotal role in advancing Japan’s drug discovery ecosystem” Alivexis Raises $5.1M USD Series D in First Close(PDF) If you’d like more information about this topic, please email at info@alivexis.com