Inference and Understanding of Immunological states via Machine Learning and Mathematical Modeling

The immune system is a biological defense which can recognize diverse pathogens and learn from the experiences of past infections. The diversity (repertoire) of adaptive immune cells, i.e., T cells and B cells, plays a fundamental role in the recognition of, and learning about, pathogens. We are trying to understand how our immune system is maintained and dynamically regulated by integrating population dynamics models and learning theories with immune repertoire sequences of immune cell populations.

Laboratory for Immune Homeostasis, RIKEN Center for Integrative Medical Sciences

• Ryo Yokota, Yuki Kaminaga and Tetsuya J. Kobayashi, Quantification of Inter-Sample Differences in T-Cell Receptor Repertoires Using Sequence-Based Information, Front. Immunol., doi: 10.3389/fimmu.2017.01500, 2017.11, https://researchmap.jp/7000018565/published_papers/10210026
•  Kazumasa B. Kaneko, Ryosuke Tateishi, Takahisa Miyao, Yuki Takakura, Nobuko Akiyama, Ryo Yokota, Taishin Akiyama & Tetsuya J. Kobayashi, Quantitative analysis reveals reciprocal regulations underlying recovery dynamics of thymocytes and thymic environment in mice, Communications Biology, volume 2, Article number: 444, doi: 10.1038/s42003-019-0688-8, 2019, https://researchmap.jp/7000018565/published_papers/24549839
• Takuya Kato and Tetsuya J. Kobayashi, Understanding adaptive immune system as reinforcement learning, Phys. Rev. Research 3, 013222, doi: 10.1103/PhysRevResearch.3.013222, 2021.3, https://researchmap.jp/7000018565/published_papers/31835527