pith. sign in

arxiv: 2606.17891 · v1 · pith:FDX5C4R7new · submitted 2026-06-16 · 🧬 q-bio.CB · math.AP· nlin.AO· physics.bio-ph

A nonlinear theory for chemotactic fronts of mixed populations

Giulia L. Celora , Marjorie Watts , Carles Falc\'o , Mohit P. Dalwadi This is my paper

Pith reviewed 2026-06-26 21:52 UTC · model grok-4.3

classification 🧬 q-bio.CB math.APnlin.AOphysics.bio-ph
keywords chemotaxisheterogeneous populationscollective migrationasymptotic analysisdensity profilesself-generated gradientsdendritic cellsT cells
0
0 comments X

The pith

Heterogeneity in diffusivity, consumption, and sensitivity produces four regimes for chemotactic density profiles in mixed cell populations.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a nonlinear theory using asymptotic analysis to describe how mixed cell populations migrate collectively under self-generated chemical gradients. It disentangles the effects of three types of heterogeneity to show that they produce exactly four distinct dynamical behaviors, covering all possible cases. This is applied to dendritic and T cells, predicting a regime with balanced mixing and front localization enabled by strong consumption creating intermediate signaling. The result gives general principles for spatial organization in heterogeneous collectives.

Core claim

Using asymptotic analysis, the full system of PDEs for heterogeneous chemotactic migration is reduced to a closed nonlinear theory. This theory reveals four distinct dynamical behaviours determined by the relative magnitudes of heterogeneity in cell diffusivity, chemoattractant consumption, and chemotactic sensitivity, which together classify all possible regimes of density profile formation.

What carries the argument

The asymptotic reduction of the PDE system to a closed nonlinear theory that yields four dynamical regimes for the density profiles.

If this is right

  • Density profiles of mixed populations can be predicted from the three heterogeneity parameters.
  • The co-migration of dendritic and T cells operates in a regime balancing mixing and T-cell localization at the front.
  • Strong consumption by dendritic cells generates intermediate long-range chemoattractant signaling.
  • All possible collective migration patterns fall into one of the four regimes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The four-regime classification could guide experiments to engineer desired spatial organizations in cell collectives.
  • Similar non-reciprocal interactions may govern other biological systems involving heterogeneous populations.
  • If validated, the theory suggests that tuning consumption rates can control the extent of mixing versus segregation.

Load-bearing premise

The asymptotic analysis produces a closed theory without residual higher-order effects altering the density profiles across all parameter combinations.

What would settle it

Finding a set of measured parameters for a cell collective whose observed density profile does not match any of the four predicted regime profiles.

Figures

Figures reproduced from arXiv: 2606.17891 by Carles Falc\'o, Giulia L. Celora, Marjorie Watts, Mohit P. Dalwadi.

Figure 1
Figure 1. Figure 1: Predictions for the coupled co-migration of T and dendritic cells. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Characterisation of consumer front morphologies. [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Diagrams illustrating how colocalization of consumer-sensor populations depends on the level of heterogeneity in the migratory properties of the two cell types for the different consumer front morphologies A-D (defined as in [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
read the original abstract

Collective migration of heterogeneous cell populations is central to many biological and physiological processes, including development and immune response. Recent experimental and theoretical advances have shown how asymmetric interactions with self-generated chemical gradients shape the spatial distribution of distinct cell types within migrating collectives. However, the principles governing robust spatial organisation of heterogeneous cell populations remain poorly understood. Here, we use asymptotic analysis to systematically derive a nonlinear analytical theory for heterogeneous cell collectives guided by self-generated chemotaxis. Our theory disentangles how heterogeneity in cell diffusivity, chemoattractant consumption, and chemotactic sensitivity shape the density profiles of migrating heterogeneous collectives, revealing four distinct dynamical behaviours that together capture all possible regimes. We calibrate our framework to experimental data on the co-migration of dendritic and T cells. We predict that this system operates in a parameter regime that balances intercellular mixing with T-cell localisation at the leading front of the migrating collective. Our theory reveals that this behaviour is enabled by intermediate long-range chemoattractant signalling generated through strong chemoattractant consumption by dendritic cells. Overall, our framework provides general principles for understanding how non-reciprocal chemical interactions shape robust collective migration in heterogeneous cell populations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript uses asymptotic analysis to derive a nonlinear analytical theory for chemotactic fronts in heterogeneous cell populations. It identifies four distinct dynamical regimes arising from heterogeneity in cell diffusivity, chemoattractant consumption, and chemotactic sensitivity, claiming these regimes together exhaustively classify all possible parameter combinations. The theory is calibrated to experimental data on dendritic and T cell co-migration and predicts that this system operates in a regime balancing intercellular mixing with T-cell localisation at the leading front, enabled by intermediate long-range signalling from strong consumption by dendritic cells.

Significance. If the asymptotic reductions are valid and the four regimes are exhaustive without residual higher-order effects, the work supplies general principles for how non-reciprocal chemical interactions organise spatial structure in mixed collectives. The calibration to immune-cell data and the concrete prediction for the dendritic-T cell system are strengths that increase the result's applicability to developmental and immunological contexts.

major comments (2)
  1. [Regime classification section] The section deriving the four regimes (likely the main results section following the asymptotic analysis): the claim that these regimes 'together capture all possible regimes' requires an explicit demonstration that the chosen scalings remain uniformly valid and that no hybrid behaviours arise at regime boundaries where competing effects enter at the same asymptotic order; without this, the exhaustiveness assertion is not yet load-bearing.
  2. [Application to dendritic-T cell data] The calibration and prediction paragraph for the dendritic-T cell system: it is unclear whether the assigned regime follows from independently measured parameter values or from fitting consumption and sensitivity to the same density profiles used for validation; this distinction is needed to confirm the prediction is falsifiable rather than post-hoc.
minor comments (2)
  1. Figure captions should explicitly indicate which of the four regimes each panel corresponds to and reference the relevant reduced equations.
  2. All asymptotic scalings and matching conditions should be numbered and cross-referenced in the text for traceability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thoughtful and constructive report. We address each major comment in detail below and have revised the manuscript to strengthen the presentation of the regime classification and the calibration procedure.

read point-by-point responses

  1. Referee: [Regime classification section] The section deriving the four regimes (likely the main results section following the asymptotic analysis): the claim that these regimes 'together capture all possible regimes' requires an explicit demonstration that the chosen scalings remain uniformly valid and that no hybrid behaviours arise at regime boundaries where competing effects enter at the same asymptotic order; without this, the exhaustiveness assertion is not yet load-bearing.

    Authors: We agree that an explicit check of uniform validity and boundary behaviour is required to make the exhaustiveness claim rigorous. In the revised manuscript we have added a new subsection (Section 3.5) together with Supplementary Note S4. There we perform a matched-asymptotic analysis at each regime boundary, retaining the next-order terms when two effects become comparable. This shows that the transition between regimes occurs smoothly at leading order and that no additional hybrid scalings appear; the four regimes therefore remain exhaustive. The added material is referenced from the main-text claim. revision: yes

  2. Referee: [Application to dendritic-T cell data] The calibration and prediction paragraph for the dendritic-T cell system: it is unclear whether the assigned regime follows from independently measured parameter values or from fitting consumption and sensitivity to the same density profiles used for validation; this distinction is needed to confirm the prediction is falsifiable rather than post-hoc.

    Authors: We have revised the relevant paragraph (now Section 4.2) to state explicitly that the consumption rates and chemotactic sensitivities are taken from independent literature values measured in separate assays (cited references), while the co-migration density profiles serve only for post-prediction validation. The regime assignment is therefore a genuine prediction based on independently determined parameter orderings. We have also added a short discussion of how future experiments could falsify the predicted regime. revision: yes

Circularity Check

0 steps flagged

Asymptotic reduction from PDEs to closed nonlinear theory is independent of target regimes and data calibration

full rationale

The derivation begins from a system of PDEs for cell densities and chemoattractant, applies asymptotic analysis with stated scalings on diffusivity, consumption, and sensitivity parameters, and obtains four reduced regimes. This reduction is performed once from the governing equations and does not invoke the experimental data or the specific dendritic-T cell prediction. Calibration occurs after the theory is derived and is used only to assign parameter values; the claim that the four regimes exhaust all combinations follows from the asymptotic balances, not from fitting. No self-citation chain, self-definitional step, or fitted quantity renamed as prediction is present in the load-bearing derivation. The framework is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the validity of the asymptotic reduction procedure and on the assumption that the three heterogeneity parameters (diffusivity, consumption rate, chemotactic sensitivity) are sufficient to classify all regimes. No explicit free parameters, axioms, or invented entities are stated in the abstract.

axioms (1)
  • domain assumption Asymptotic analysis yields a closed nonlinear theory whose four regimes exhaust all possible combinations of the three heterogeneity parameters
    Invoked by the statement that the theory 'reveals four distinct dynamical behaviours that together capture all possible regimes'

pith-pipeline@v0.9.1-grok · 5753 in / 1433 out tokens · 22926 ms · 2026-06-26T21:52:38.321630+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

47 extracted references · 38 canonical work pages · 1 internal anchor

  1. [1]

    K. J. Cheung and S. Horne-Badovinac. Collective cell mi- gration modes in development, tissue repair and cancer. Nature Reviews Molecular Cell Biology, 26(10):741–758,

  2. [2]

    doi:10.1038/s41580-025-00858-9

    work page doi:10.1038/s41580-025-00858-9

  3. [3]

    Wu and M

    Z. Wu and M. Wong. Collective cell migration across scales: A systems perspective.Seminars in Cell & Developmental Biology, 173:103628, 2025. doi:10.1016/j.semcdb.2025.103628

    work page doi:10.1016/j.semcdb.2025.103628 2025

  4. [4]

    S. J. Stehbens, E. Scarpa, et al. Perspectives in collective cell migration – moving forward.Journal of Cell Science, 137(12):jcs261549, 2024. doi:10.1242/jcs.261549

    work page doi:10.1242/jcs.261549 2024

  5. [5]

    Alanko, M

    J. Alanko, M. C. Uçar, et al. CCR7 acts as both a sensor and a sink for CCL19 to coordinate collective leukocyte migration.Science Immunology, 8(87):eadc9584, 2023. doi:10.1126/sciimmunol.adc9584

    work page doi:10.1126/sciimmunol.adc9584 2023

  6. [6]

    E. Donà, J. D. Barry, et al. Directional tissue migration through a self-generated chemokine gradient.Nature, 503(7475):285–289, 2013. doi:10.1038/nature12635

    work page doi:10.1038/nature12635 2013

  7. [7]

    H. Z. Ford, G. L. Celora, et al. Pattern forma- tion along signaling gradients driven by active droplet behavior of cell swarms.Proceedings of the Na- tional Academy of Sciences, 122(21):e2419152122, 2025. doi:10.1073/pnas.2419152122

    work page doi:10.1073/pnas.2419152122 2025

  8. [8]

    A. Kiran,P. A. Thomason, et al. Macrophage chemotaxis steered by complex self-generated gradients of comple- ment C5a.bioRxiv, 2025. doi:10.1101/2025.11.05.686842. 9

    work page doi:10.1101/2025.11.05.686842 2025

  9. [9]

    Malet-Engra, W

    G. Malet-Engra, W. Yu, et al. Collective Cell Motil- ity Promotes Chemotactic Prowess and Resistance to Chemorepulsion.Current Biology, 25(2):242–250, 2015. doi:10.1016/j.cub.2014.11.030

    work page doi:10.1016/j.cub.2014.11.030 2015

  10. [10]

    Tweedy, P

    L. Tweedy, P. A. Thomason, et al. Seeing around corners: Cells solve mazes and respond at a distance using at- tractant breakdown.Science, 369(6507):eaay9792, 2020. doi:10.1126/science.aay9792

    work page doi:10.1126/science.aay9792 2020

  11. [11]

    M. C. Uçar, Z. Alsberga, et al. Self-generated chemotaxis of mixed cell populations.Proceed- ings of the National Academy of Sciences of the United States of America, 122(34):e2504064122, 2025. doi:10.1073/pnas.2504064122

    work page doi:10.1073/pnas.2504064122 2025

  12. [12]

    Tweedy and R

    L. Tweedy and R. H. Insall. Self-Generated Gradi- ents Yield Exceptionally Robust Steering Cues.Fron- tiers in Cell and Developmental Biology, 8:133, 2020. doi:10.3389/fcell.2020.00133

    work page doi:10.3389/fcell.2020.00133 2020

  13. [13]

    McLennan, L

    R. McLennan, L. Dyson, et al. Multiscale mecha- nisms of cell migration during development: theory and experiment.Development, 139(16):2935–2944, 2012. doi:10.1242/dev.081471

    work page doi:10.1242/dev.081471 2012

  14. [14]

    X. Fu, S. Kato, et al. Spatial self-organization re- solves conflicts between individuality and collective migration.Nature Communications, 9(1):2177, 2018. doi:10.1038/s41467-018-04539-4

    work page doi:10.1038/s41467-018-04539-4 2018

  15. [15]

    E. F. Keller and L. A. Segel. Traveling bands of chemo- tactic bacteria: A theoretical analysis.Journal of Theo- retical Biology, 30(2):235–248, 1971. doi:10.1016/0022- 5193(71)90051-8

    work page doi:10.1016/0022- 1971

  16. [16]

    W. J. M. Ridgway, M. P. Dalwadi, et al. Motility- Induced Phase Separation Mediated by Bacterial Quo- rum Sensing.Physical Review Letters, 131(22):228302,

  17. [17]

    doi:10.1103/PhysRevLett.131.228302

    work page doi:10.1103/physrevlett.131.228302

  18. [18]

    M. M. Salek, F. Carrara, et al. Bacterial chemotaxis in a microfluidic T-maze reveals strong phenotypic hetero- geneity in chemotactic sensitivity.Nature Communica- tions, 10:1877, 2019. doi:10.1038/s41467-019-09521-2

    work page doi:10.1038/s41467-019-09521-2 2019

  19. [19]

    H. H. Mattingly and T. Emonet. Collective behavior andnongeneticinheritanceallowbacterialpopulationsto adapt to changing environments.Proceedings of the Na- tional Academy of Sciences, 119(26):e2117377119, 2022. doi:10.1073/pnas.2117377119

    work page doi:10.1073/pnas.2117377119 2022

  20. [20]

    S. Gude, E. Pinçe, et al. Bacterial coexistence driven by motility and spatial competition.Nature, 578(7796):588– 592, 2020. doi:10.1038/s41586-020-2033-2

    work page doi:10.1038/s41586-020-2033-2 2020

  21. [21]

    Y. Bai, C. He, et al. Spatial modulation of individual behaviors enables an ordered structure of diverse pheno- types during bacterial group migration.eLife, 10:e67316,

  22. [22]

    doi:10.7554/eLife.67316

    work page doi:10.7554/elife.67316

  23. [23]

    F. R. Macfarlane, T. Lorenzi, et al. The impact of phe- notypic heterogeneity on chemotactic self-organisation. Bulletin of Mathematical Biology, 84(12):143, 2022. doi:10.1007/s11538-022-01099-z

    work page doi:10.1007/s11538-022-01099-z 2022

  24. [24]

    Phenotype structuring in collective cell migration: a tu- torial of mathematical models and methods

    T. Lorenzi, K. J. Painter, et al. Phenotype structuring in collective cell migration: a tutorial of mathematical models and methods.Journal of Mathematical Biology, 90(6):61, 2025. doi:10.1007/s00285-025-02223-y

    work page doi:10.1007/s00285-025-02223-y 2025

  25. [25]

    Freingruber, T

    V. Freingruber, T. Lorenzi, et al. Trait-structured chemo- taxis: exploring ligand-receptor dynamics and travelling wave properties in a Keller–Segel model.Nonlinearity, 38(10):105006, 2025. doi:10.1088/1361-6544/ae03ff

    work page doi:10.1088/1361-6544/ae03ff 2025

  26. [26]

    Watts, C

    M. Watts, C. Falco, et al. Patterns of collective migration in heterogeneous, self-generated chemotaxis. [Manuscript in preparation], 2026

    2026

  27. [27]

    L. C. Evans.Partial differential equations, volume 19 of Graduate studies in mathematics. American Mathemati- cal Society, second edition, 2010

    2010

  28. [28]

    M. J. Ablowitz and A. Zeppetella. Explicit solu- tions of Fisher’s equation for a special wave speed. Bulletin of Mathematical Biology, 41(6):835–840, 1979. doi:10.1007/BF02462380

    work page doi:10.1007/bf02462380 1979

  29. [29]

    C. M. Bender and S. A. Orszag.Advanced Mathematical Methods for Scientists and Engineers I. Springer, New York, NY, 1999

    1999

  30. [30]

    J. Canosa. On a nonlinear diffusion equation describing population growth.IBM Journal of Research and Devel- opment, 17(4):307–313, 1973. doi:10.1147/rd.174.0307

    work page doi:10.1147/rd.174.0307 1973

  31. [31]

    D. D. Patel, W. Koopmann, et al. Chemokines have diverse abilities to form solid phase gra- dients.Clinical Immunology, 99(1):43–52, 2001. doi:10.1006/clim.2000.4997

    work page doi:10.1006/clim.2000.4997 2001

  32. [32]

    T. V. Phan, H. H. Mattingly, et al. Direct measurement of dynamic attractant gradients reveals breakdown of the Patlak–Keller–Segel chemotaxis model.Proceedings of the National Academy of Sciences, 121(3):e2309251121,

  33. [33]

    doi:10.1073/pnas.2309251121

    work page doi:10.1073/pnas.2309251121

  34. [34]

    G. L. Celora, B. J. Walker, et al. Chemotaxis of cell ag- gregates: morphology and dynamics of migrating active droplets.arXiv, 2026. doi:10.48550/arXiv.2602.20088

    work page doi:10.48550/arxiv.2602.20088 2026

  35. [35]

    T. J. Jewell, S. W. S. Johnson, et al. Cell-cell adhesion cannot sustain extended follower streams in a minimal non-local model of leader-follower migration.arXiv, 2026. doi:10.48550/arXiv.2604.15283

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2604.15283 2026

  36. [36]

    Lawson-Keister and M

    E. Lawson-Keister and M. L. Manning. Collective chemotaxis in a Voronoi model for confluent clus- ters.Biophysical Journal, 121(23):4624–4634, 2022. doi:10.1016/j.bpj.2022.10.029. 10

    work page doi:10.1016/j.bpj.2022.10.029 2022

  37. [37]

    Rossetti, S

    L. Rossetti, S. Grosser, et al. Optogenetic generation of leader cells reveals a force–velocity relation for collective cell migration.Nature Physics, 20(10):1659–1669, 2024. doi:10.1038/s41567-024-02600-2

    work page doi:10.1038/s41567-024-02600-2 2024

  38. [38]

    D. P. Panigrahi, G. L. Celora, et al. Intermit- tent attachments form three-dimensional cell aggre- gates with emergent fluid properties.bioRxiv, 2025. doi:10.1101/2025.09.24.678186

    work page doi:10.1101/2025.09.24.678186 2025

  39. [39]

    Dolak and C

    Y. Dolak and C. Schmeiser. Kinetic models for chemo- taxis: Hydrodynamic limits and spatio-temporal mecha- nisms.Journal of Mathematical Biology, 51(6):595–615,

  40. [40]

    doi:10.1007/s00285-005-0334-6

    work page doi:10.1007/s00285-005-0334-6

  41. [41]

    Saragosti, V

    J. Saragosti, V. Calvez, et al. Mathematical De- scription of Bacterial Traveling Pulses.PLOS Computational Biology, 6(8):e1000890, 2010. doi:10.1371/journal.pcbi.1000890

    work page doi:10.1371/journal.pcbi.1000890 2010

  42. [42]

    R. Alert. Cellular Sensing Governs the Stability of Chemotactic Fronts.Physical Review Letters, 128(14),

  43. [43]

    doi:10.1103/PhysRevLett.128.148101

    work page doi:10.1103/physrevlett.128.148101

  44. [44]

    Bhattacharjee, D

    T. Bhattacharjee, D. B. Amchin, et al. Chemotactic smoothing of collective migration.eLife, 11:e71226, 2022. doi:10.7554/eLife.71226

    work page doi:10.7554/elife.71226 2022

  45. [45]

    T. E. Fjelde, K. Xu, et al. Turing.jl: a general-purpose probabilistic programming language.ACM Transac- tions on Probabilistic Machine Learning, 1(3), 2025. doi:10.1145/3711897

    work page doi:10.1145/3711897 2025

  46. [46]

    M. D. Hoffman and A. Gelman. The No-U-Turn sam- pler: adaptively setting path lengths in Hamiltonian Monte Carlo.Journal of Machine Learning Research, 15(1):1593–1623, 2014. doi:10.5555/2627435.2638586

    work page doi:10.5555/2627435.2638586 2014

  47. [47]

    DifferentialEquations.jl – A Performant and Feature-Rich Ecosystem for Solving Differential Equations in Julia.Journal of Open Research Software, 5(1):15, May 2017

    C. Rackauckas and Q. Nie. DifferentialEquations.jl–a performant and feature-rich ecosystem for solving dif- ferential equations in Julia.Journal of Open Research Software, 5(1), 2017. doi:10.5334/jors.151. 11 Supplementary Information Contents A Details of the SC model 1 A.1 Simulation of the dendritic/T-cell system . . . . . . . . . . . . . . . . . . . ....

    work page doi:10.5334/jors.151 2017