A machine learning approach to discover migration modes and transition dynamics of heterogeneous dendritic cells

Aug 5, 2026

Upload by Logic

Dendritic cell migration is essential for initiating immune responses, yet its highly heterogeneous motility and mode switching remain poorly characterized. This study applies an unsupervised machine learning framework to long-term two-dimensional migration trajectories from GM-CSF-derived bone-marrow derived dendritic cells. Three migration modes emerge—slow-diffusive, slow-persistent, and fast-persistent—independent of cellular state. Immature cells exhibit frequent mode changes with a history-dependent transition pattern, while mature cells show no directional bias.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Introduction

Background on dendritic cell motility

Limitations of average dichotomous descriptions

Study hypothesis and unsupervised ML strategy

TYPE Original Research PUBLISHED 04 April 2023 DOI 10.3389/fimmu.2023.1129600 OPEN ACCESS EDITED BY Manfred B. Lutz, Julius Maximilian University of Würzburg, Germany REVIEWED BY Junsang Doh, Seoul National University, Republic of Korea Guillaume Darrasse-Jeze, Universite´ de Paris, France *CORRESPONDENCE Jae-Hyung Jeon [jeonjh@postech.ac.kr](jeonjh@postech.ac.kr)[ ](jeonjh@postech.ac.kr)Yoon-Kyoung Cho [ykcho@unist.ac.kr](ykcho@unist.ac.kr) †These authors have contributed equally to this work SPECIALTY SECTION This article was submitted to Antigen Presenting Cell Biology, a section of the journal Frontiers in Immunology RECEIVED 22 December 2022 ACCEPTED 06 March 2023 PUBLISHED 04 April 2023 CITATION Song T, Choi Y, Jeon J-H and Cho Y-K (2023) A machine learning approach to discover migration modes and transition dynamics of heterogeneous dendritic cells. Front. Immunol. 14:1129600 . doi: 10.3389/fimmu.2023.1129600 COPYRIGHT © 2023 Song, Choi, Jeon and Cho. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. A machine learning approach to discover migration modesand transition dynamics of heterogeneous dendritic cells Taegeun Song 1,2†, Yongjun Choi 3,4†, Jae-Hyung Jeon 1,5* and Yoon-Kyoung Cho 3,4* 1 Department of Physics, Pohang University of Science and Technology (POSTECH), Pohang, Republic of Korea, 2 Department of Data information and Physics, Kongju National University, Gongju, Republic of Korea, 3Center for Soft and Living Matter, Institute for Basic Science (IBS), Ulsan, Republic of Korea, 4 Department of Biomedical Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, Republic of Korea, 5Asia Paciﬁc Center for Theoretical Physics (APCTP), Pohang, Republic of Korea Dendritic cell (DC) migration is crucial for mounting immune responses. Immature DCs (imDCs) reportedly sense infections, while mature DCs (mDCs) move quickly to lymph nodes to deliver antigens to T cells. However, their highly heterogeneous and complex innate motility remains elusive. Here, we used an unsupervised machine learning (ML) approach to analyze long-term, twodimensional migration trajectories of Granulocyte-macrophage colonystimulating factor (GMCSF)-derived bone marrow-derived DCs (BMDCs) . We discovered three migratory modes independent of the cell state: slow-diffusive (SD), slow-persistent (SP), and fast-persistent (FP) . Remarkably, imDCs more frequently changed their modes, predominantly following a u nicyclic SD! FP! SP! SD transition, whereas mDCs showed no transition directionality. We report that DC migration exhibits a history-dependent mode transition and maturation-dependent motility changes are emergent properties of the dynamic switching of the three migratory modes. Our ML-based investigation provides new insights into studying complex cellular migratory behavior. KEYWORDS dendritic cell, cell migration, machine learning, transition dynamics, maturation 1 Introduction Cell migration is essential for homeostasis in living systems (1) . Intriguingly, cell motility shows complex dynamics beyond the classical diffusion theory (2) . Therefore, various random-walk models have been employed to explain anomalous diffusion processes (3, 4) . For instance, bacterial micro-swimmers and T cells deploy an effective intermittent search process, alternating between slow and fast motion, such as the run-andtumble motion and Ĺevy walk (5, 6) . Frontiers in Immunology 01 [frontiersin.org](frontiersin.org) Dendritic cells (DCs) exhibit adaptive motility patterns, reﬂec

FAQs

What method is used to study dendritic cell migration in this work?

An unsupervised machine learning approach is applied to long-term two-dimensional single-cell migration trajectories.

How many migration modes are identified, and what are they?

Three modes are discovered: slow-diffusive (SD), slow-persistent (SP), and fast-persistent (FP).

Do immature and mature dendritic cells show different transition behaviors?

Yes. Immature dendritic cells change modes more frequently and show a predominant SD→FP→SP→SD transition directionality, while mature cells show no transition directionality.

Aug 5, 2026

Upload by Logic