Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Self-organized dynamics and emergent shape spaces of active isotropic fluid surfaces

T0 review · 3 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper establishes a variational formulation of active isotropic fluid surfaces: the stationary points of a dissipation functional built from entropy production, supplemented by Lagrange multipliers that enforce geometric constraints…

desk verdict A genuinely useful variational method for active surface shape spaces, with an overbroad invertibility claim that needs a caveat but doesn't break the main results. read the letter →

arxiv 2501.17849 v2 pith:6LSW5NYM submitted 2025-01-29 cond-mat.soft

Da Gao , Huayang Sun , Rui Ma , Alexander Mietke This is my paper · ORCID
classification cond-mat.soft
keywords activefluidsurfacesdissipationfunctionalOnsagerrelationsRayleighshapespaceanalysismembranemechanicshydrodynamicscreeningvariationalmethod
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper claims that the shape dynamics of active fluid surfaces—minimal models for membranes, cell cortices, and tissues—can be recast as a variational problem. The authors construct a Rayleigh dissipation functional from the entropy production of the surface and show that, after changing the thermodynamic flux-force pairing, even reactive (non-dissipative) active couplings contribute to the functional. The stationary points of the constrained functional are shown to yield exactly the geometries and flows that satisfy force and torque balance for the prescribed linear constitutive laws. This equivalence turns the search for stationary shapes into a boundary value problem, so emergent shape spaces can be computed directly rather than only explored by time-stepping dynamics. The paper demonstrates the framework on open membranes and closed active surfaces, revealing first-order shape transitions, degenerate solution branches, and a role for hydrodynamic screening in division-like geometry.

What carries the argument

The load-bearing object is the constrained Rayleigh functional $\bar{R}=R+L$, where $R$ is assembled from the free-energy change rate and half the total entropy production, and $L$ uses Lagrange multipliers $\alpha,\beta,\zeta$ to impose the geometric identities $r'=h\cos\psi$, $z'=-h\sin\psi$, and the scaling Lagrangian-Eulerian (SLE) parameterization. In the SLE parameterization, fixed mesh coordinates map to physical arc length through a single global scale factor $h(t)$, so numerical resolution is controlled independently of local flows. Variation of $\bar{R}$ with respect to velocities and shape-derivative fields produces a coupled system of first-order ordinary differential equations whose boundary terms become the boundary conditions of the surface problem. The identity that makes the construction work is the flux-force transformation (35)–(36): choosing $r_p$ instead of $\Delta\mu$ as the thermodynamic force makes the active cross-coupling $\xi/\Lambda$ appear with the same sign in both constitutive laws, so it contributes to dissipation and the entropy production becomes a valid variational principle.

What would settle it

Take a linear constitutive law with a positive semidefinite but singular Onsager matrix—for example, Eqs. (27) and (33) with $\Lambda=0$ and $\xi\neq 0$—and compare the direct force-balance solution for an axisymmetric surface with the stationary points of any candidate Rayleigh functional; if the variational equations fail to reproduce the same geometry and flow, or if the flux-force transformation violates the second law, the claimed equivalence is false.

Watch

Extended reading notes

Core claim

The central claim is an equivalence: the functional $R(v,r_p)=dF/dt + D(v,r_p)$ in Eq. (39), together with the Lagrange-multiplier terms $L$ in Eq. (43), has stationary points that are precisely the surfaces and flows satisfying force balance $\mathrm{div}(T)=-f^{\mathrm{ext}}$ and torque balance $\mathrm{div}(M)=-\epsilon:T$ with the tension $T=T_e+T_d$ and moment $M=M_e$ defined by the constitutive laws (27)–(29). The equivalence is verified explicitly for axisymmetric surfaces in Appendix E of the paper. The crucial step is replacing the chemical potential difference $\Delta\mu$ by the reaction rate $r_p$ as the thermodynamic force, which turns the reactive Onsager coefficient $\xi$ into a dissipative coupling so that the entropy production defines a genuine Rayleigh functional. As a result, stationary geometries and flows of active surfaces can be computed directly as solutions of a boundary value problem, without first simulating the full time-dependent dynamics.

Load-bearing premise

The construction presupposes that the Onsager coefficient matrix is invertible and that exchanging $\Delta\mu$ for the reaction rate $r_p$ as the thermodynamic force preserves the second law, so every linear constitutive law can be recast in purely dissipative form and a Rayleigh functional exists.

Editorial extensions

If this is right

  • Stationary geometries and flows of an active fluid surface can be obtained directly from a boundary value problem, without time-stepping through transients.
  • Nonlinear shape spaces can be mapped on dense parameter grids, exposing Gibbs loops, disconnected solution branches, and first-order transitions between protrusion geometries.
  • For closed mechanochemically active surfaces, the framework predicts multi-stable stationary shape regions and a reentrant symmetry-breaking and restoring instability controlled by hydrodynamic screening.
  • Hydrodynamic screening, set by the ratio $R_0/L_h$, controls whether contractile rings produce sharply divided dumbbells or broader, more elongated ingressions during division-like transformations.
  • The same variational construction applies to other linear constitutive laws for active surfaces with different broken symmetries, providing a route to classify their stationary shape spaces.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: the variational equivalence may extend beyond axisymmetry, but the analog of the SLE parameterization would need to control shearing and twisting of the surface map, not just meridional stretching; the paper identifies this as an open question.
  • Editorial inference: if the dissipation functional is convex near a stationary solution, its second variation could supply a direct linear-stability criterion, complementing the perturb-and-simulate stability tests used in the paper.
  • Editorial inference: the first-order transitions and hysteresis found in the protrusion and division models suggest that biological systems could exploit bistable shape spaces for irreversible developmental decisions, a functional speculation the paper does not pursue.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The manuscript introduces a variational (Rayleigh-functional) formulation for active isotropic fluid surfaces. Starting from the internal entropy production of previous surface-hydrodynamics frameworks, the authors rewrite the linear Onsager constitutive laws by exchanging the thermodynamic force Δμ for the reaction rate r_p, obtaining a dissipation functional R whose stationary points are claimed to reproduce force and torque balance, including reactive non-dissipative couplings. Geometric relations and a scaling Lagrangian-Eulerian (SLE) parameterization are imposed through Lagrange multipliers, reducing stationary shape and flow computation to boundary value problems. The method is applied to open Helfrich-type membranes with spontaneous-curvature coatings and pulling forces, and to closed active surfaces with concentration-dependent active tensions, yielding phase diagrams for protrusion, polarization/propagation, and guided division. Appendix E contains an explicit derivation of the equivalence between the variational equations and force balance for axisymmetric surfaces.

Significance. If correct, the construction provides a systematic way to generate variational principles for active surface theories, extending the equilibrium membrane variational approach to non-equilibrium settings while retaining reactive couplings. The paper is strong on concreteness: Appendix E gives a detailed axisymmetric derivation, the numerical code is released on GitHub, and the applications are tested against known results such as the Derenyi et al. tube-pulling force and previous mechanochemical polarization instabilities. The dense parameter sweeps and explicit stability checks add substantial value. The main caveat is that the flux-force transformation underlying the functional is not as general as claimed, and the general (non-axisymmetric) equivalence is not proven; these issues affect the advertised scope of the method but not the specific numerical phase diagrams.

major comments (3)
  1. [Sec. IV D, Eqs. (35)–(36)] The transformation from the (v, Δμ) to the (v, r_p) description divides by the coefficient Λ. The paper states that Onsager relations guarantee an invertible coefficient matrix, but positive-semidefinite symmetric matrices are not generally invertible, and invertibility of the full matrix is also not sufficient for this specific swap: for Λ=0 and ξ≠0 the constitutive matrix [[η_b, ξ],[-ξ, Λ]] has determinant η_bΛ+ξ² > 0, yet Eqs. (35)–(36) are undefined because Λ appears in the denominator. The existence of the Rayleigh functional therefore requires Λ>0 (and η_b, η_s ≥ 0 for convexity), not merely Onsager reciprocity. Since the numerical applications prescribe ξΔμ(c) and never use r_p or Λ, the phase diagrams are unaffected, but the advertised generality for "the full set of a priori defined constitutive laws" is overclaimed. Please either impose and state the condition Λ>0 or show how singular/diffusionless cases are to be handled.
  2. [Sec. IV D and App. E] The central equivalence claim is proven only for axisymmetric surfaces. Appendix E derives the Euler-Lagrange equations for the meridional, azimuthal, and normal variations in the axisymmetric setting. The unconstrained dissipation functional is stated to be valid for arbitrary surfaces, but no general tensor derivation of δR/δv = 0 from Eqs. (20)–(21) is given. If the general statement is to be retained, a coordinate-free or general-coordinate proof should be added; otherwise the claim should be qualified to axisymmetric surfaces. This does not affect the examples, all of which are axisymmetric, but it matters for the advertised scope of the method.
  3. [Sec. IV F 2 and Sec. V B] The paper does not specify how the stationary concentration field c(u) is determined in the direct stationary computations that produce Figs. 5 and 7. The variation (45) is over velocities and geometric time-derivatives, not over c, and the reaction-diffusion dynamics (52) is not listed among the ODEs in Table II. If the steady-state concentration is obtained by appending Eq. (52) with ∂_t c = 0 to the boundary value problem, that should be stated explicitly; otherwise the phase diagrams cannot be reproduced from the text alone.
minor comments (4)
  1. [Sec. IV D] The sentence "One can directly verify that if coefficients η, η_b, ξ are such that the second law is respected in the original entropy production ... then it will also be respected by Θ̂int" is true but not demonstrated; adding the explicit identity Θ̂int(v, r_p) = Θint(v, Δμ), which follows by substituting Eqs. (27),(33) into (32) and Eqs. (35),(36) into (37), would remove the appearance of an unproved assertion.
  2. [Sec. V A 2] The claimed quantitative agreement with ref. [35] appears arithmetically inconsistent: with γ0 = 12.5 κ/R_b², the formula f0 = 2π√(γ0κ) gives f0 ≈ 22.2 κ/R_b, so κ/R_b ≈ f0/22 rather than approximately f0/5. Please clarify the comparison or correct the stated factor.
  3. [Table III] In the ∂th row it is unclear whether the boundary conditions ζ(0)=ζ(1)=0 apply to both open and closed surfaces; the text should state explicitly what replaces these conditions for closed surfaces, where no physical boundary exists.
  4. [App. E] The notation switches between v_u and normalized components such as ̄v_u without always recalling that barred components refer to the normalized basis ̄e_u = e_u/h; a brief notational remark at the start of Appendix E would improve readability.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the variational equivalence is verified algebraically in App. E; self-citations are not load-bearing, and the main caveat (Onsager invertibility) is a correctness/scope issue, not a circular reduction.

full rationale

The central derivation is self-contained. Constitutive laws (27)-(29) are stated as inputs; the flux/force swap in Eqs. (35)-(36) is an algebraic inversion of the input laws (requiring Lambda != 0); and the dissipation functional (39) is built from the free-energy rate (31), the internal entropy production (37), and external dissipation (34). Appendix E explicitly proves the claimed equivalence: tangential variations give the meridional and azimuthal force balances (E30) and (E34), normal variations give n . div(Td + Te) = 0 for both Eulerian (E40) and SLE (E52) parameterizations, and delta R / delta rp = 0 returns the input rp law as a consistency condition rather than as an independent prediction. Numerical applications prescribe the active stress xi Delta mu(c) directly and involve no fitting of the quantities being predicted; the comparison with the Derenyi tube-pulling threshold is an external benchmark, not a calibration. Self-citations, notably refs. [19,20], are used to supply the active-surface model and the linear instability threshold Pe*, which are independent inputs and do not justify the variational equivalence. The one substantive weakness is the claim in Sec. IV D that 'Onsager relations guarantee an invertible coefficient matrix': positive-semidefinite Onsager matrices need not be invertible, and the specific inversion divides by Lambda, which can vanish. This is a mathematically overbroad generality claim and a correctness/scope caveat, but it does not amount to a circular reduction of the paper's predictions to its inputs; the worked examples do not use rp or Lambda.

Assumptions & free parameters 9 free parameters · 6 assumptions · 0 invented entities

No new physical entities are introduced; the Lagrange multipliers and the SLE coordinate flow are mathematical devices, not new physical degrees of freedom. The model parameters are chosen by hand for illustration, not fitted to data. The central axioms are Onsager linear irreversible thermodynamics and the overdamped surface mechanics.

free parameters (9)
  • Active contractility (xi*Delta mu)_0 = 450 D*eta_b/R0^2
    Sets the strength of active stress in closed surface examples; chosen far above the linear instability threshold. Not fitted to data.
  • Friction coefficient Gamma = 0.09 eta_b/R0^2 (closed), 0.25 eta_s/Rb^2 (open)
    Sets hydrodynamic screening length; varied to produce Figs. 5 and 7. Chosen by hand.
  • Turnover rate k = 45 tau_eta^-1 (default), 9 tau_eta^-1 in Fig. 5
    Controls restoration of preferred concentration profile; varied for stability diagram Fig. 7c.
  • Saturation concentration c_s = 10 c0 (default), 1 c0 in Fig. 5
    Softens active stress saturation; affects critical Peclet number.
  • Passive tension gamma = 9 D*eta_b/R0^2
    Renormalized isotropic tension; chosen to keep spherical state stable.
  • Diffusion constant D = 1 R0^2/tau_eta
    Sets Peclet number scale in the concentration dynamics.
  • Spontaneous curvature profile parameters C0_hat, sigma_p, A_p = C0_hat*Rb = 9, sigma_p = 25, A_p = 2/25 (2 pi Rb^2)
    Model the protein coating patch in open membrane examples; profile shape chosen by hand.
  • Pulling force profile parameters f0, sigma_f, A_f = f0 varies, sigma_f = 250, A_f = 2/250 (2 pi Rb^2)
    Model external force patch in open membrane examples.
  • Reference concentration profile width w = 0.02
    Width of the contractile ring in Eq. (53).
assumptions (6)
  • standard math Onsager reciprocal relations and linear irreversible thermodynamics
    Justifies the constitutive laws Eqs. (27) and (33) and the flux-force transformation in Sec. IV C-D.
  • standard math Surface differential geometry identities
    Used throughout App. A-B for metric, curvature, and divergence expressions.
  • domain assumption Overdamped dynamics (inertia neglected)
    Stated in Sec. III A: momentum balance uses no inertial term, appropriate for biological cells and tissues.
  • ad hoc to paper Linear constitutive laws with invertible Onsager matrix
    Needed to ensure existence of an equivalent all-dissipative representation; paper overstates Onsager guarantees as discussed in red flags.
  • domain assumption Single-field concentration model in applications
    Sec. V B uses one concentration c with reaction-diffusion dynamics Eq. (52) and active stress Eq. (54), rather than the two-species fuel/product model in Sec. IV.
  • domain assumption Constant chemical potential difference Delta mu in the thermodynamic derivation
    Sec. IV B assumes external flux maintains Delta mu constant; this simplifies the free energy dynamics and is not carried into the concentration-field examples.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Self-organized dynamics and emergent shape spaces of active isotropic fluid surfaces." pith.science (2026). https://pith.science/paper/6LSW5NYM

@misc{pith2026250117849,
  author       = {Pith},
  title        = {Pith review of: Self-organized dynamics and emergent shape spaces of active isotropic fluid surfaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6LSW5NYM}},
  note         = {Machine review of arXiv:2501.17849}
}
read the original abstract

Theories of self-organized active fluid surfaces have emerged as an important class of minimal models for the shape dynamics of biological membranes, cells and tissues. However, due to their inherent geometric nonlinearities and the absence of general minimization principles in active systems, it remains a major challenge to systematically study the emergent shape spaces such theories give rise to. Here, we introduce a novel variational approach that allows for a direct computation of stationary surface geometries and flows, which enables the classification of non-equilibrium phase transitions in shape spaces described by active surface theories. To achieve this, we construct a dissipation functional systematically from the entropy production in active surfaces and show how generic symmetries imposed by Onsager relations can be exploited to also account for reactive non-dissipative terms in constitutive laws. This functional is supplemented by Lagrange multipliers that relax nonlinear geometric constraints, which leads to a tractable variational problem suitable for implicit dynamic simulations and explicit calculations of non-trivial steady state geometries and flows. We apply this framework to study the dynamics of open fluid membranes and closed active fluid surfaces, and characterize the space of stationary solutions that corresponding surfaces and flows occupy. These analyses rationalize the interplay of first-order shape transitions of internally and externally forced fluid membranes, reveal degenerate regions in stationary shape spaces of mechanochemically active surfaces and identify a mechanism by which hydrodynamic screening controls the geometry of active surfaces undergoing cell division-like shape transformations.

Figures

Figures reproduced from arXiv: 2501.17849 by the authors.

Figure 1
Figure 1. Parameterization of open (a) and closed (b) axisymmetric surfaces [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Dynamics of an open membrane with local spontaneous curvature coating [yellow patches, see Eq. (49)]. [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Nonlinear shape space analysis of open membranes. [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Self-organization dynamics, contractile ring slipping and propagation of a closed active isotropic fluid surface. [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Shape space characterization of polarized stationary surfaces. [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: Dynamics of guided furrow formation and emergence of extreme localized curvature (Movie 4). [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 7
Figure 7. Figure 7: Shape space analysis of guided division model with parameters used in Fig. 6 indicated by green circles. [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Coupling between Phase Separation and Geometry on a Closed Elastic Curve: Free Energy Minimization and Dynamics

    cond-mat.soft 2026-02 conditional novelty 6.0 of 10

    Closure of a deformable elastic loop changes phase separation, producing stable and metastable peanut/polygon multi-domain states absent on rigid or open substrates.

Reference graph

Works this paper leans on

87 extracted references · 79 canonical work pages · cited by 1 Pith paper

  1. [1]

    II the right-hand side contains the time derivatives ∂tr, ∂tz, ∂tψ, which enter all equations by construction linearly

    Implicit time-stepping method for the surface dynamics For the ODEs listed in Tab. II the right-hand side contains the time derivatives ∂tr, ∂tz, ∂tψ, which enter all equations by construction linearly. We can therefore design a fully im- plicit time integration scheme to evolve the surface shape by replacing ∂tr = r(u, t) − r(u, t− ∆t) ∆t ∂tz = z(u, t) −...

  2. [2]

    Direct computation of stationary geometries and flows While dynamic simulations are useful to explore the parameter-dependence of a given model, they are not practical to systematically characterize the space of non-trivial station- ary surface geometries and flows or the bifurcation structure of this space. With the formulation derived above, we can dire...

  3. [3]

    2a, inset), as can be induced by, for example, protein-patches on lipid membranes that play a role in clathrin-mediated endocytosis [37, 38, 57]

    Local spontaneous curvature We first assign a finite spontaneous curvature “coating” to a sub-region of the membrane (Fig. 2a, inset), as can be induced by, for example, protein-patches on lipid membranes that play a role in clathrin-mediated endocytosis [37, 38, 57]. We as- sume this coating occupies the same material points at all the times. Starting fr...

  4. [4]

    The external force density amounts 10 Figure 3

    Pulling forces We next study membrane deformations under an external force density f ext = f ext z ez, while maintaining the sponta- neous curvature coating. The external force density amounts 10 Figure 3. Nonlinear shape space analysis of open membranes.(a) Height ˜d = d/Rb of stationary surfaces as function of spontaneous curvature ˜C0 = ˆC0Rb in a fini...

  5. [5]

    (53) for which the reference concentration ˆc0 = c0 in Eq

    Spontaneous polarization and propagation We first discuss the case ∆c = 0 in Eq. (53) for which the reference concentration ˆc0 = c0 in Eq. (52) is constant. Using this model, it has been shown in the absence of friction that a spherical surface with uniform distribution of stress-regulator Parameter Value Unit κ Bending rigidity 1 κ ηb Bulk viscosity 1 η...

  6. [6]

    (53)] with an elevated recruitment rate around the equator ( kon = kˆc0)

    Guided division We finally characterize surface shapes that emerge when turnover of the stress regulator is spatially inhomogeneous [∆c > 0, see Eq. (53)] with an elevated recruitment rate around the equator ( kon = kˆc0). A representative example of the shape dynamics is depicted in Fig. 6. We set cs = 10c0 in Eq. (54) to avoid saturating local active te...

  7. [7]

    Kaksonen and A

    M. Kaksonen and A. Roux, Nat. Rev. Mol. Cell Biol. 19, 313 (2018)

    2018

  8. [8]

    Jahn and T

    R. Jahn and T. C. S¨udhof, Annu. Rev. Biochem.68, 863 (1999)

    work page 1999

Show all 87 references
  1. [9]

    L.-G. Wu, E. Hamid, W. Shin, and H.-C. Chiang, Annu. Rev. Physiol. 76, 301 (2014)

    2014

  2. [10]

    Alberts, D

    B. Alberts, D. Bray, J. Lewis, M. Raff, K. Roberts, and J. Wat- son, Molecular Biology of the Cell, 4th ed. (Garland, 2002)

    2002

  3. [11]

    Phillips, J

    R. Phillips, J. Kondev, J. Theriot, and H. Garcia, Physical biol- ogy of the cell (Garland Science, 2012)

    2012

  4. [12]

    Dimova, S

    R. Dimova, S. Aranda, N. Bezlyepkina, V . Nikolov, K. A. Riske, and R. Lipowsky, J. Phys. Condens. Matter 18, S1151 (2006)

    2006

  5. [13]

    Salbreux, G

    G. Salbreux, G. Charras, and E. Paluch, Trends Cell Biol. 22, 536 (2012)

    2012

  6. [14]

    Bray and J

    D. Bray and J. White, Science 239, 883 (1988)

    1988

  7. [15]

    Mayer, M

    M. Mayer, M. Depken, J. S. Bois, F. J ¨ulicher, and S. W. Grill, Nature 467, 617 (2010)

    2010

  8. [16]

    Lecuit and P.-F

    T. Lecuit and P.-F. Lenne, Nat. Rev. Mol. Cell Biol. 8, 633 (2007)

    2007

  9. [17]

    Chugh, A

    P. Chugh, A. G. Clark, M. B. Smith, D. A. D. Cassani, K. Dierkes, A. Ragab, P. P. Roux, G. Charras, G. Salbreux, and E. K. Paluch, Nat. Cell Biol. 19, 689 (2017)

    2017

  10. [18]

    Turlier, B

    H. Turlier, B. Audoly, J. Prost, and J.-F. Joanny, Biophys. J. 106, 114 (2014)

    2014

  11. [19]

    A. C. Martin, M. Gelbart, R. Fernandez-Gonzalez, M. Kaschube, and E. F. Wieschaus, J. Cell Biol. 188, 735 (2010)

    2010

  12. [20]

    Heisenberg and Y

    C.-P. Heisenberg and Y . Bella¨ıche, Cell 153, 948 (2013)

    2013

  13. [21]

    Ranft, M

    J. Ranft, M. Basan, J. Elgeti, J.-F. Joanny, J. Prost, and F. J¨ulicher, Proc. Natl. Acad. Sci. U.S.A. 107, 20863 (2010)

    2010

  14. [22]

    Daeden, A

    A. Daeden, A. Mietke, E. Derivery, C. Seum, F. J ¨ulicher, and M. Gonzalez-Gaitan, Nat. Cell Biol. 25, 235 (2023)

    2023

  15. [23]

    Denk-Lobnig, J

    M. Denk-Lobnig, J. F. Totz, N. C. Heer, J. Dunkel, and A. C. Martin, Development 148, dev199232 (2021)

    2021

  16. [24]

    Salbreux and F

    G. Salbreux and F. J ¨ulicher, Phys. Rev. E96, 032404 (2017)

    2017

  17. [25]

    Mietke, F

    A. Mietke, F. J ¨ulicher, and I. F. Sbalzarini, Proc. Natl. Acad. Sci. U.S.A. 116, 29 (2019)

    2019

  18. [26]

    Mietke, V

    A. Mietke, V . Jemseena, K. V . Kumar, I. F. Sbalzarini, and F. J¨ulicher, Phys. Rev. Lett.123, 188101 (2019)

    2019

  19. [27]

    Torres-S ´anchez, M

    A. Torres-S ´anchez, M. K. Winter, and G. Salbreux, PLOS Comp. Biol. 18, e1010762 (2022)

    2022

  20. [28]

    Khoromskaia and G

    D. Khoromskaia and G. Salbreux, eLife 12, e75878 (2023)

    2023

  21. [29]

    Bailleul, N

    R. Bailleul, N. Cuny, D. Khoromoskaia, S. Basu, G. Bergamini, P. Cucurachi, S. Rupp, A. Guse, C. Curantz, N. Swinhoe, P. Cleves, J. Craggs, S. Fujita, Y .-i. Nakajima, P. J. Steenber- gen, A. Diz-Mu˜noz, G. Salbreux, and A. Ikmi, bioRxiv (2025)

    2025

  22. [30]

    P. B. Canham, J. Theor. Biol. 26, 61 (1970)

    1970

  23. [31]

    Helfrich, Zeitschrift f ¨ur Naturforschung C 28, 693 (1973)

    W. Helfrich, Zeitschrift f ¨ur Naturforschung C 28, 693 (1973)

    1973

  24. [32]

    E. A. Evans, Biophys. J. 14, 923 (1974)

    1974

  25. [33]

    Zhong-can and W

    O.-Y . Zhong-can and W. Helfrich, Phys. Rev. Lett. 59, 2486 (1987). 16

    1987

  26. [34]

    Zhong-can and W

    O.-Y . Zhong-can and W. Helfrich, Phys. Rev. A 39, 5280 (1989)

    1989

  27. [35]

    Lipowsky, Nature 349, 475 (1991)

    R. Lipowsky, Nature 349, 475 (1991)

    1991

  28. [36]

    Seifert, K

    U. Seifert, K. Berndl, and R. Lipowsky, Phys. Rev. A 44, 1182 (1991)

    1991

  29. [37]

    Jian-Guo and O.-Y

    H. Jian-Guo and O.-Y . Zhong-Can, Phys. Rev. E47, 461 (1993)

    1993

  30. [38]

    J ¨ulicher and U

    F. J ¨ulicher and U. Seifert, Phys. Rev. E 49, 4728 (1994)

    1994

  31. [39]

    Lipowsky, Biophys

    R. Lipowsky, Biophys. J. 64, 1133 (1993)

    1993

  32. [40]

    J ¨ulicher and R

    F. J ¨ulicher and R. Lipowsky, Phys. Rev. E53, 2670 (1996)

    1996

  33. [41]

    Der ´enyi, F

    I. Der ´enyi, F. J¨ulicher, and J. Prost, Phys. Rev. Lett.88, 238101 (2002)

    2002

  34. [42]

    Seifert, Adv

    U. Seifert, Adv. Phys. 46, 13 (1997)

    1997

  35. [43]

    Hassinger, G

    J. Hassinger, G. Oster, D. Drubin, and P. Rangamani, Biophys. J. 112, 310a (2017)

    2017

  36. [44]

    Walani, J

    N. Walani, J. Torres, and A. Agrawal, Proc. Natl. Acad. Sci. U.S.A. 112, E1423 (2015)

    2015

  37. [45]

    L. V . Chernomordik and M. M. Kozlov, Nat. Struct. Mol. Biol. 15, 675 (2008)

    2008

  38. [46]

    Arroyo and A

    M. Arroyo and A. DeSimone, Phys. Rev. E 79, 031915 (2009)

    2009

  39. [47]

    Doi, Journal of Physics: Condensed Matter 23, 284118 (2011)

    M. Doi, Journal of Physics: Condensed Matter 23, 284118 (2011)

    2011

  40. [48]

    Arroyo, N

    M. Arroyo, N. Walani, A. Torres-S ´anchez, and D. Kaurin, in The Role of Mechanics in the Study of Lipid Bilayers (Springer International Publishing, 2018) pp. 287–332

    2018

  41. [49]

    Salbreux, F

    G. Salbreux, F. J ¨ulicher, J. Prost, and A. Callan-Jones, Phys. Rev. Res. 4, 033158 (2022)

    2022

  42. [50]

    Katchalsky and P

    A. Katchalsky and P. F. Curran,Nonequilibrium Thermodynam- ics in Biophysics (Harvard University Press, 1965)

    1965

  43. [51]

    Onsager, Phys

    L. Onsager, Phys. Rev. 37, 405 (1931)

    1931

  44. [52]

    Onsager, Phys

    L. Onsager, Phys. Rev. 38, 2265 (1931)

    1931

  45. [53]

    J ¨ulicher, S

    F. J ¨ulicher, S. W. Grill, and G. Salbreux, Rep. Prog. Phys. 81, 076601 (2018)

    2018

  46. [54]

    Torres-S ´anchez, D

    A. Torres-S ´anchez, D. Mill´an, and M. Arroyo, Journal of Fluid Mechanics 872, 218’ ¨A`ı271 (2019)

    2019

  47. [55]

    A. Sahu, Y . A. Omar, R. A. Sauer, and K. K. Mandadapu, J Comp. Phys. 407, 109253 (2020)

    2020

  48. [56]

    L. D. Wittwer and S. Aland, Journal of Computational Physics: X 17, 100126 (2023)

    2023

  49. [57]

    See www.github.com/amietke/ActiveShapeSpaces

  50. [58]

    C. Hirt, A. Amsden, and J. Cook, J. Comp. Phys. 14, 227 (1974)

    1974

  51. [59]

    J ¨ulicher, A

    F. J ¨ulicher, A. Ajdari, and J. Prost, Rev. Mod. Phys. 69, 1269 (1997)

    1997

  52. [60]

    Prost, F

    J. Prost, F. J ¨ulicher, and J. F. Joanny, Nat. Phys.11, 111 (2015)

    2015

  53. [61]

    Meixner, Ann

    J. Meixner, Ann. Phys. 43, 244 (1943)

    1943

  54. [62]

    Kierzenka and L

    J. Kierzenka and L. F. Shampine, ACM Trans. Math. Softw.27, 299 (2001)

    2001

  55. [63]

    Ma and J

    R. Ma and J. Berro, Biophysical Journal 120, 52a (2021)

    2021

  56. [64]

    Agrawal and D

    A. Agrawal and D. J. Steigmann, Biomechanics and Modeling in Mechanobiology 8, 371 (2009)

    2009

  57. [65]

    J. S. Bois, F. J ¨ulicher, and S. W. Grill, Phys. Rev. Lett. 106, 028103 (2011)

    2011

  58. [66]

    Bonati, L

    M. Bonati, L. D. Wittwer, S. Aland, and E. Fischer-Friedrich, New Journal of Physics 24, 073044 (2022)

    2022

  59. [67]

    Th ´ery, A

    M. Th ´ery, A. Jim´enez-Dalmaroni, V . Racine, M. Bornens, and F. J¨ulicher, Nature 447, 493 (2007)

    2007

  60. [68]

    N. T. L. Rodrigues, S. Lekomtsev, S. Jananji, J. Kriston-Vizi, G. R. X. Hickson, and B. Baum, Nature 524, 489 (2015)

    2015

  61. [69]

    Sedzinski, M

    J. Sedzinski, M. Biro, A. Oswald, J.-Y . Tinevez, G. Salbreux, and E. Paluch, Nature 476, 462 (2011)

    2011

  62. [70]

    M. J. Lighthill, Commun. Pure Appl. Math. 5, 109 (1952)

    1952

  63. [71]

    Lauga and T

    E. Lauga and T. R. Powers, Reports on Progress in Physics 72, 096601 (2009)

    2009

  64. [72]

    Schmitt and H

    M. Schmitt and H. Stark, Physics of Fluids 28, 012106 (2016)

    2016

  65. [73]

    Z ¨ottl and H

    A. Z ¨ottl and H. Stark, Journal of Physics: Condensed Matter 28, 253001 (2016)

    2016

  66. [74]

    C. A. Whitfield and R. J. Hawkins, New J. Phys. 18, 123016 (2016)

    2016

  67. [75]

    L. D. Wittwer and S. Aland, J. Comp. Phys. X 17, 100126 (2023)

    2023

  68. [76]

    Kreyszig, Introduction to differential geometry and Rieman- nian geometry (University of Toronto Press, 1968)

    E. Kreyszig, Introduction to differential geometry and Rieman- nian geometry (University of Toronto Press, 1968)

    1968

  69. [77]

    A. Sahu, R. A. Sauer, and K. K. Mandadapu, Phys. Rev. E 96, 042409 (2017). 17 Appendix A: Differential geometry of axisymmetric surfaces

    2017

  70. [78]

    Geometric surface properties For axisymmetric surfaces described by X(u, ϕ, t) given in Eq. (8), tangent vectors ei = ∂iX and surface normal n = eu × eϕ/|eu × eϕ| are given by eu = h   cos ψ cos ϕ cos ψ sin ϕ − sin ψ   eϕ = r   − sin ϕ cos ϕ 0   n =   sin ψ cos ϕ sin...

  71. [79]

    It is convenient to use normalized tangent basis vectors¯eu := eu/h and ¯eϕ := eϕ/r, with ei given in Eqs

    Vector and tensor divergence We provide for reference explicit expressions for the divergences div(v) = ei · ∂iv = ∇ivi + C k k vn div(T) = ei · ∂iT = ∇iti (A12) of a vector and of a tension tensor field v and T, respectively, on an axisymmetric surface parameterized by mesh c...

  72. [80]

    (C1) for an ALE-parameterized surface, we have to approximate (denoting for brevity ˜f = f √g) F (t + δt) = Z ω(t+δt) ˜f (¯s1, ¯s2, t+ δt)d¯s1d¯s2, (C4) to linear order in δt

    Moving boundary integrals with ALE parameterization To determine the total time derivative ofF (t) in Eq. (C1) for an ALE-parameterized surface, we have to approximate (denoting for brevity ˜f = f √g) F (t + δt) = Z ω(t+δt) ˜f (¯s1, ¯s2, t+ δt)d¯s1d¯s2, (C4) to linear order in...

  73. [81]

    sin ψ r − C0 2 − ψ′ h 2# ¯eu ⊗ ¯eu + κ 2

    Generalized conservation laws and continuity equation If F (t) is a conserved quantity, we must have dF dt = 0 and the right-hand side of Eq. (C10) can be rearranged into ALE: Z Ω(t) ∂t (f √g) ds1ds2 = − I ∂Ω(t) f qiνids, (C14) where we have used the covariant Stokes theorem [...

  74. [82]

    (40) in the main text reads R(v, rp) = ∂tfκ + fκ ∂t(rh) rh − rp∆µ + J ext n − f ext · v + γdiv(v) + 1 2 (S1 + S2 + S3) + 1 2 Γv2 ∥ rh, (E14) where fκ is given in Eq

    Dissipation functional density on an axsymmetric surface The density of the Rayleigh dissipation functional in the (v, rp) ensemble defined by Eq. (40) in the main text reads R(v, rp) = ∂tfκ + fκ ∂t(rh) rh − rp∆µ + J ext n − f ext · v + γdiv(v) + 1 2 (S1 + S2 + S3) + 1 2 Γv2 ∥...

  75. [83]

    Relevance of different parameterization choices for the variation When discussing variations in terms of a conventional Euler parameterization ( qi = vi), we substitute the time derivatives in Eqs. (E14) and (E18) by ∂tr = vn sin ψ (E20) ∂tz = vn cos ψ (E21) ∂th = ψ′vn (E22) ∂...

  76. [84]

    (E14) 1 h ηb( ¯Suu + ¯Sϕϕ) + ηs( ¯Suu − ¯Sϕϕ) + γ + ξ∆µ ′ + 2ηs( ¯Suu − ¯Sϕϕ) cos ψ r − Γ¯vu + f ext u + ∂fκ ∂C0 C ′ 0 h = 0, (E28) where we have used Eq

    Tangential force balance from variational equations The Euler-Lagrange equation for variations with respect to meridional flows, 1 2π δR δ¯vu = ∂R ∂¯vu − ∂R ∂¯v′u ′ = 0, (E27) implies for R(v, rp) given in Eq. (E14) 1 h ηb( ¯Suu + ¯Sϕϕ) + ηs( ¯Suu − ¯Sϕϕ) + γ + ξ∆µ ′ + 2ηs( ¯S...

  77. [85]

    E 2) and are therefore discussed separately in the following two sections

    Normal force balance from variational equation The variational steps that recover the normal force balance equation are distinct for Eulerian parameterization and the SLE parameterization (see Sec. E 2) and are therefore discussed separately in the following two sections. a. U...

  78. [86]

    (40) contains a non-trivial boundary term that generally modifies boundary terms of the variation

    Boundary conditions with Eulerian parameterization Note that the Rayleigh functional Eq. (40) contains a non-trivial boundary term that generally modifies boundary terms of the variation. For the variations computed in Sec. E 4 a they read ∂R ∂[(¯vu)′] + fH r δ¯vu u=1 u=0 = 2π...

  79. [87]

    Boundary conditions with SLE parameterization and geometric constraints For the variations computed in Sec. E 4 b they now read ∂ ¯R ∂[(¯vu)′] + ∂(fH qurh) ∂¯vu δ¯vu u=1 u=0 = 2πr ¯eu · (fH G + Td) · ¯euδ¯vu u=1 u=0 (F5) ∂ ¯R ∂[(¯vϕ)′] δ¯vϕ u=1 u=0 = 2πr ¯eu · (fH G + Td) · ¯e...

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.