HOME / Exhibition Program / Seeking to capture the growth trajectory of cells
Exhibition Program
Machine Learning Science
01

Seeking to capture the growth trajectory of cells

Interactive hypothesis testing on cell differentiation

Seeking to capture the growth trajectory of cells
Abstract

Biological cells develop into various organs and tissues through differentiation. How individual cells undergo this differentiation process is a critical and universal question in the life sciences. To elucidate these differentiation mechanisms, it is essential to both generate hypotheses and test them. Therefore, we are constructing a new mathematical framework for hypothesis generation and testing. Specifically, this technology assists life science researchers in interactively discovering hypotheses through AI, which aids in the visual design of differentiation structure hypotheses and in testing of their validity. Deeper insights into the mechanisms of biological cell differentiation may enable us to induce differentiation phenomena involving various genetic and environmental factors. By investigating the causes of congenital diseases and the efficacy of medications for designated intractable diseases, we aim to contribute to the advancement of regenerative medicine and artificial organ technology.

Seeking to capture the growth trajectory of cells
References

[1] M. Nakano, H. Sakuma, R. Nishikimi, K. Komiya, T. Iwata, K. Kashino, “HyperbolicPHATE: Visualizing Continuous Hierarchy of Latent Differentiation Structures,” IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2025.

Poster
Contact

Masahiro Nakano, Biomedical Informatics Research Group, Media Information Laboratory

Click here for other research exhibits