Kiyoshi Asai

Optimal design of mRNA by deep generative models using biological information archive

Research Director

Kiyoshi Asai

Collaborator

Hiroshi Abe	Professor Graduate School of Science Nagoya University
Yutaka Saito	Professor School of Frontier Engineering Kitasato University
Kengo Sato	Professor School of System Design and Technology Tokyo Denki University

Outline

We will develop a technology to design mRNAs with desired performance from protein amino acid sequences by utilizing the archive of publicly available biological information and deep learning, and by using a DBTL cycle that greatly reduces the amount of experimentation. Proteins to be designed are selected from medical applications such as viral vaccines, cancer vaccines, gene editing, and protein replacement therapy. Effectiveness of the technology will be demonstrated by measurement of the performance of the designed mRNAs.

Kaoru Inokuchi

Functional analyses of the sleeping brain driven by multi-layered neural activity data

Research Director

Kaoru Inokuchi

Collaborator

Taro Toyoizumi	Team Leader Center for Brain Science RIKEN
Tatsuya Haga	Tenure-track researcher Advanced ICT Research Institute National Institute of Information and Communications Technology

Outline

We aim to apply a large language model to the multi-layered, multi-dimensional data obtained from the brain during learning and sleep. This will lead to the development of a data analysis system that visualizes crucial neural activities during sleep. We will establish a feedback loop between experiments, data science, and mathematical models. This will uncover the mechanisms of information processing in both non-REM and REM sleep, shedding light on the brain’s learning processes during sleep. Additionally, we will propose functional hypotheses regarding sleep-related brain learning and construct mathematical models to implement them.

Hiroshi Ochiai

Understanding the mammalian cell fate regulation framework through multimodal spatiotemporal integrated omics analysis

Research Director

Hiroshi Ochiai

Collaborator

Soya Shinkai	Senior Scientist Center for Biosystems Dynamics Research RIKEN
Takuya Funatomi	Associate professor Graduate School of Science and Technology Nara Institute of Science and Technology
Kazumitsu Maehara	Assistant Professor Medical Institute of Bioregulation Kyushu University

Outline

The process of mammalian development involves numerous stochastic and unknown factors, making a detailed understanding of the cell fate regulation mechanism challenging. In this research, we will target the early developmental stages of mice and organoids to analyze the cell-to-cell heterogeneity in gene expression, the epigenomic factors that regulate it, and intercellular interactions in a data-driven, multimodal, and spatiotemporal manner, elucidating the cell fate regulation mechanism.

Kumi Kuroda

Future childcare: Wearable AI solutions for infant sleep and crying

Research Director

Kumi Kuroda

Collaborator

Natsue Yoshimura

Professor School of Computing Tokyo Institute of Technology

Outline

Excessive crying in infants can stress parents. In this research, we will develop a sleep-support wearable system that utilizes the transport response, in which the child becomes docile when carried. Then, from experimental data at home, we will construct a machine-learning model that estimates the infant’s probability of falling asleep after childcare activities from pulse wave information before the activities. From this model, we will evaluate the autonomic nervous system activities that control sleep onset and verify this model in physiological experiments on mice, to control infant night crying and predict sleep onset.

Hirokazu Toju

Multistability of microbiomes: multidisciplinary science for the optimization of ecosystem-level functions

Research Director

Hirokazu Toju

Collaborator

Kenta Suzuki	Research & Development Scientist BioResource Research Center RIKEN
Masato Yamamichi	Associate Professor National Institute of Genetics Research Organization of Information and Systems

Outline

We will establish techniques to overview the “multistability” of microbiomes from the perspective of ecosystem-level biological functions. In particular, we will focus on the phenomenon of rapid changes in species compositions, elucidating the biological and non-biological core factors driving the time-series dynamics of ecosystems based on the integration of causal inference and network science. Furthermore, we will construct experimental systems for plant- and fish-associated microbiomes, providing general technical frameworks for controlling ecosystem-level dynamics and functions in the medical, agricultural, and industrial fields

Hideto Mori

Automating DNA Design & Synthesis Using Conversational AI

Research Director

Hideto Mori

Outline

Advancements in genome editing and synthetic biology have amplified the demand for DNA synthesis, underscoring the urgent need for more sophisticated automation in this field. However, adopting automated processes is constrained by the complexities of operating such systems and the absence of established protocols for sharing experimental procedures. In this study, we aim to address these challenges by developing a conversational AI agent that utilizes advanced large-scale language models to facilitate DNA design and synthesis automation.