​Topic

​This session of the Tokyo AI (TAI) Advanced AI (AAI) sub-group will be on topics in Bio x AI, bringing together experts in the field to talk about how AI is transforming biological research and applications. We'll cover AI-driven advancements in drug discovery, biomanufacturing, and systems biology, bridging the gaps between machine learning, healthcare, and biotechnology.

​​Our Community

​​​​Tokyo AI (TAI) is a community composed of people based in Tokyo and working with, studying, or investing in AI. We are engineers, product managers, entrepreneurs, academics, and investors intending to build a strong “AI coreˮ in Tokyo. Find more in our overview: https://bit.ly/tai_overview

​Schedule

​17:30 - 18:00 Doors Open

​18:00 - 18:05 Introduction

​18:05 - 18:50 Creating an Engine for Scientific Discovery: Unlocking the World of Biology with AI - Sucheendra Kumar (profile)

​18:50 - 19:10 Building a High-Abstraction AI Development Environment in AI Drug Discovery - Keisuke Kamata (profile)

​19:10 - 20:00 Break/Networking/Food

​20:00 - 20:30 Unlocking the Plant Kingdom: Revolutionizing Biomanufacturing x AI, through Plant Cell Agriculture - Paula Elbl (profile)

​20:30 - 21:00 Data-Driven Drug Discovery: Axcelead’s AI-Powered Platform, BAP - Kent Kawata (profile)

​21:00 Wrap-up

​Speakers

​Sucheendra Kumar (profile)

​Title: Creating an Engine for Scientific Discovery: Unlocking the World of Biology with AI

​Abstract:

​Biological systems are inherently complex, driven by interactions across multiple layers of organization, from genomics to metabolomics. This complexity is magnified by the exponential growth of high-throughput biological data, creating challenges in interpretation and discovery. Recent advancements in artificial intelligence, particularly transformer-based models and other foundational architectures offer new capabilities to navigate this complexity and accelerate biomedical research. This talk will begin by exploring how the community is adapting these AI models to address the challenges posed by biology’s multi-layered complexity. We will then present two specific examples from our own work that aim to build towards the vision of creating an engine for scientific discovery. The first focuses on literature-driven hypothesis generation to identify novel hypotheses by analyzing scientific literature at scale. The second highlights the development of multiomics foundation models that integrate diverse biological datasets, enabling a deeper understanding of complex systems and new avenues for therapeutic development. By combining the power of these models with the context of biology, we can uncover insights that were previously inaccessible, paving the way for a transformative approach to biomedicine.

​Bio:

​Dr. Sucheendra K. Palaniappan (Suchee) leads the Data Science and Engineering teams at SBX Corporation, Tokyo, and is a Scientist at the Systems Biology Institute, Tokyo. His work focuses on the intersection of technology and biomedicine, where he develops AI-driven platforms and solutions to address challenges in precision medicine and healthcare delivery. By integrating computational approaches with deep biological context, his team creates systems that bridge foundational science with practical biomedical applications. He holds a PhD in Computer Science from the School of Computing, National University of Singapore.

​Keisuke Kamata (profile)

​Title: Building a High-Abstraction AI Development Environment in the Increasingly Complex and Advanced Field of AI Drug Discovery

​Abstract:

​The advent of Transformer-based models has significantly advanced AI-driven drug discovery. However, training these models requires substantial GPU resources and advanced knowledge in AI implementations beyond biology, raising the technical barrier to entry. To lower this barrier, NVIDIA's BioNeMo facilitates the distributed learning of specialized models in biology and biochemistry across multiple nodes. It simplifies the pre-training and fine-tuning of protein language models, making it widely accepted among researchers. When utilizing BioNeMo, an appropriate management workflow to systematically build assets is also necessary. Weights & Biases plays a crucial role here, as it allows for the simple organizational management of model training processes and version control of datasets and models. This tool is used by many engineers worldwide, and the integration of BioNeMo with Weights & Biases has led to the development of a user-friendly AI integration platform. A speaker from Weights & Biases, who contributed to the recent November 2024 release of BioNeMo2, will discuss the extent of abstraction achieved with current tools and what is critical for researchers advancing AI drug discovery.

​Bio:

​Keisuke earned his master's degree by analyzing data from animal experiments he conducted during his university years. Since then, he has been dedicatedly working in the fields of life sciences and AI. As an AI solutions engineer skilled in causal inference, ML, and LLMs, Keisuke has served as the healthcare lead at DataRobot and currently at Weights & Biases, supporting the AI adoption of numerous companies within the healthcare industry. During the COVID-19 pandemic, he collaborated with the National Center for Global Health and Medicine to produce research papers and letters to the Ministry of Health, Labour and Welfare. He has also contributed to the development of BioNeMo2, a framework used by many researchers in AI drug discovery.

​Paula Elbl (profile)

​Title: Unlocking the Plant Kingdom: Revolutionizing Biomanufacturing x AI, through Plant Cell Agriculture

​Abstract:

​The potential of plant-based compounds remains largely untapped, with over 97% of the Plant Kingdom’s bioactives yet to be explored. Paula Elbl, Co-Founder and CSO of Cyanotype Bio, shares how AI-driven platforms are revolutionizing R&D and product discovery in plant biotechnology. This talk will dive into how Cyanotype Bio integrates advanced metabolomics and artificial intelligence to identify, optimize, and scale sustainable plant-based ingredients for industries such as cosmetics, agriculture, and botanical drugs. Paula will also showcase her company's unique approach to conservation and scalable biomanufacturing, using AI to unlock the full potential of biodiversity while addressing global sustainability challenges.

​Bio:

​Dr. Paula Elbl is the co-founder and CSO of Cyanotype Bio, a plant biotechnology company that applies AI-driven innovation to the advanced biomanufacturing of plant-based bioactives. Growing up in Amazonia inspired her deep-rooted passion for plant biodiversity and conservation, leading to a PhD in Botany and a multidisciplinary postdoc at the University of São Paulo. With 18+ years of expertise in plant cell agriculture, Paula’s work spans artificial seed development to biomanufacturing novel materials. She has received accolades including the Woman Founder in Biotechnology Award and the Global Change Award And the H&M Foundation Global Change Award for lab-grown cotton technology through her previous start-up, GALY Co.

​Kent Kawata (profile)

​Title: Data-Driven Drug Discovery: Axcelead’s AI-Powered Platform, BAP

​Abstract:

​Axcelead, Japan’s premier healthcare platform company, leverages a world-class drug discovery platform inherited from a leading global pharmaceutical company. With a rich legacy of over 1,000 projects, an extensive compound library, and cutting-edge AI capabilities, we tackle the inherent complexity of drug discovery. Our proprietary platform, Beyond A Platform (BAP), integrates high-quality legacy data, industry-leading AI models, and a seamless lab-in-the-loop system to deliver superior outcomes. In this talk, we will explore how Axcelead combines advanced generative AI and domain expertise to overcome challenges in drug discovery. We’ll share insights into how BAP enhances screening efficiency, predictive modeling, and compound generation, ultimately accelerating the journey from hit identification to lead optimization.

​Bio:

​Kent serves as the Operations and BD Lead at the Digital Unit of Axcelead Drug Discovery Partner. With a background in biotech, materials modeling, and computational chemistry, he has held leadership roles spanning R&D, new business development, and the startup ecosystem. He began his career at Asahi Kasei Corporation’s analytics lab and spent the majority of his career at the German company Merck, where he spearheaded and commercialized cross-functional innovation projects in life sciences. More recently, Kent has been deeply involved in biotech startups, including serving as CEO of venture companies.

​​Organizers

​Gergely (Greg) Juhasz: Associate Professor at the Institute of Science Tokyo (formerly Tokyo Tech). His expertise lies in computational chemistry, and he is a member of the TAC-MI program, which bridges Chemical and Materials Science with Data Science and Machine Learning. He earned his PhD in Materials Chemistry at Kyushu University and has research experience at Carnegie Mellon University, Kyushu University, and Tokyo Tech. His research focuses on quantum chemical and computational approaches to problems in nanocarbons, catalysis, and electrochemical systems.

​​Ilya Kulyatin: Fintech and AI entrepreneur with work and academic experience in the US, Netherlands, Singapore, UK, and Japan, with an MSc in Machine Learning from UCL.