Pith. sign in

REVIEW 1 major objections 4 minor 38 references

Feedback-mediated circulation and persistence of stochastic fluctuations in gene regulatory circuits

T0 review · 1 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read This paper claims that in two-node gene regulatory motifs, total node-wise noise decomposes into intrinsic, extrinsic, and a new feedback-generated cyclic component whose sign—set by the product of the two regulatory sensitivities—determine

desk verdict Solid LNA decomposition with a useful cyclic-noise term, but the temporal persistence claim is undercut by the wrong no-feedback baseline—worth reading, needs revision. read the letter →

arxiv 2607.17743 v1 pith:XOP66VTF submitted 2026-07-20 physics.bio-ph q-bio.MN

Nashita Rahman , Mintu Nandi , Sudip Chattopadhyay , Suman K Banik This is my paper · ORCID
classification physics.bio-phq-bio.MN
keywords cyclicnoisefeedbackgainlinearapproximationtwo-nodemotifsdecompositiongeneregulatorycircuitsautocorrelationLyapunovequation
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

The paper tries to establish that feedback loops in the simplest two-node gene circuits do more than rescale existing noise: they inject a distinct, additive noise component caused purely by loop closure. This 'cyclic noise' is positive for reinforcing motifs (mutual activation or mutual repression), amplifying node-wise fluctuations, and negative for opposing motifs (one activation, one repression), attenuating them. The same mechanism leaves a temporal fingerprint: reinforcing loops slow the decay of autocorrelation, while opposing loops leave it close to the no-feedback baseline. The decomposition follows analytically from the linear noise approximation and is checked against stochastic simulation. If correct, it gives an experimental route to read feedback sign and strength from measured means, variances, covariances, and autocorrelations.

What carries the argument

The load-bearing object is the Lyapunov equation JΣ + ΣJᵀ + 2D = 0 for the steady-state covariance matrix Σ, with the 2×2 Jacobian J containing the two regulatory sensitivities f′_XY and f′_YX. Solving this equation analytically yields the three-term noise decomposition. The new term, cyclic noise, is controlled by the dimensionless feedback gain H = f′_XY f′_YX/(β_X β_Y); its sign sets whether loop closure adds to or subtracts from node-wise noise, and H = 1 marks the stability boundary for reinforcing motifs. The temporal analysis relies on the short-time expansion C(τ) ≈ (I + Jτ)Σ, which separates the birth-death relaxation 1 − β_N τ from the feedback-coupled autocorrelation term.

What would settle it

Measure the total squared coefficient of variation of a node in a synthetic two-node motif with known sign of H, then disconnect the feedback loop (e.g., by removing the second regulator) while keeping the direct regulation and degradation rates fixed. If total noise does not increase when H>0 or decrease when H<0 relative to the open-loop reference, the cyclic-noise term as stated is wrong. More directly, compute H from independently measured regulatory sensitivities and check whether the extracted cyclic-noise sign from variance data matches the sign of H.

Watch

Extended reading notes

Core claim

For each transcription factor N in a two-node feedback motif, the paper argues that the squared coefficient of variation obeys η_N² = η_int,N² + η_ext,N² + η_cyc,N², with cyclic noise η_cyc,N² = T_MN H/(1−H) η_int,N², where H = f′_XY f′_YX/(β_X β_Y) is the dimensionless feedback gain and T_MN = β_M/(β_N+β_M) is a time-averaging factor. H is positive for reinforcing motifs (++ and −−), making cyclic noise positive and divergent as H→1−; H is negative for opposing motifs (+− and −+), making cyclic noise negative. The paper further derives the short-lag autocorrelation ρ_NN(τ) = 1 − β_N τ + τ f′_NM (⟨n_M⟩/⟨n_N⟩) η²_MN/η²_N, where the feedback-dependent term is large enough in reinforcing motifs

Load-bearing premise

The linear noise approximation is the load-bearing premise: the covariance decomposition and the cyclic-noise sign rely on fluctuations staying small enough that the linearized Lyapunov equation accurately describes the steady state, so the results fail near bistability, bursting, or large-amplitude oscillations.

Editorial extensions

If this is right

  • If the decomposition is correct, measuring a node's mean copy number, variance, and covariance with its partner suffices to extract cyclic noise experimentally from single-cell time-series data.
  • Reinforcing feedback motifs (++ and −−) will show amplified node noise and longer-lived autocorrelation, providing a quantitative signature of positive-feedback-driven fluctuation persistence even below the bistability threshold.
  • Opposing motifs (+− and −+) will show reduced node noise and autocorrelation decay close to the no-feedback baseline, offering a way to identify negative feedback from noise statistics alone.
  • The divergence of cyclic noise as H→1− predicts that strong reinforcing feedback can produce large, slow fluctuations before any bifurcation, a precursor that could be observed in synthetic toggle circuits.
  • Because the formulas are expressed in measurable quantities, the framework can be applied directly to existing single-cell reporter data from synthetic gene circuits without additional fitting assumptions.

Reading between the lines

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

  • Editorial extension: for networks with more than two nodes, the same Lyapunov-based decomposition should generalize to a sum over closed loops, with each loop contributing a term proportional to its own feedback gain; this would make cyclic noise a graph-theoretic quantity.
  • Editorial extension: the negative cyclic noise in opposing motifs resembles a noise-cancellation term; it may connect to control-theoretic limits on noise suppression, suggesting that opposing feedback achieves attenuation by redirecting fluctuations rather than by adding dissipation.
  • Editorial extension: a direct testable prediction beyond the paper is that in a synthetic (+−) motif, the total squared coefficient of variation should fall below the value obtained by breaking the loop (open-loop cascade) while keeping all other parameters fixed; this would confirm the negative cyclic contribution.
  • Editorial extension: the autocorrelation slowdown in reinforcing motifs implies that positive feedback lengthens the effective memory of a gene circuit, which could be exploited in synthetic memory devices and may be observable in cell-fate decision circuits before commitment.
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

1 major / 4 minor

Summary. This manuscript develops a linear-noise-approximation (LNA) framework for two-node gene-regulatory feedback motifs and proposes an additive decomposition of node-wise noise into intrinsic, extrinsic, and a new 'cyclic' component. The cyclic component, η_cyc,N = T_MN H/(1−H) η_int,N, is positive for reinforcing motifs and negative for opposing motifs, and the authors verify the analytic expressions with stochastic simulations. The paper further claims a temporal signature of feedback circulation based on a short-time expansion of the steady-state autocorrelation, arguing that reinforcing loop closure prolongs fluctuation persistence relative to a no-feedback reference. The steady-state decomposition is clean and analytically well grounded; the temporal interpretation is less well controlled.

Significance. If the steady-state decomposition stands, it gives a compact and interpretable topological readout: the sign of the feedback gain H determines whether loop closure amplifies or attenuates node-wise fluctuations, and the explicit formula for η_cyc is a useful analytic result. The derivation is self-contained, the definitions are clear, and the simulation agreement is a genuine strength. The temporal-signature claim, however, is not yet supported by the chosen comparison, so the present version overstates one of its advertised contributions. With the temporal claim either corrected or appropriately softened, the paper is a solid contribution to the noise-decomposition literature for feedback motifs.

major comments (1)
  1. [§III.C, Eqs. (10)–(11), Figs. 3–4] The 'no-feedback reference' is taken as ρ_NN^(0)(τ)=1−β_N τ, the autocorrelation of an isolated birth–death process. This does not control for the direct regulatory edge in the two-node motif. Equation (11) shows that the feedback-mediated correction actually contains both extrinsic noise propagation and cyclic noise: ρ_NN(τ)=1−β_N τ + τ f'_NM... (η^2_MN/η^2_N). The second term is nonzero even in an open-loop cascade with the same direct edge and H=0, because upstream fluctuations propagate through that edge. Therefore, the deviations from ρ_NN^(0) seen in Figs. 3 and 4 cannot be attributed specifically to loop closure, and the conclusion that 'positive cyclic noise is associated with ... longer temporal persistence' is not supported by the chosen comparison. To isolate the temporal signature of cyclic circulation, the authors should compare the closed-loop motif against the open-loop ca
minor comments (4)
  1. [Fig. 4 caption] The caption says 'panels (e)-(f)' for the Y autocorrelations; this should be '(e)-(h)'.
  2. [Fig. 3 caption] Typo: 'The quantitie ρ_NN^(0)' should read 'The quantities ρ_NN^(0)'.
  3. [Eq. (11) and Appendix C] The notation η^2_MN is used in the main text before it is explicitly defined. Define it as the normalized covariance σ^2_{MN}/⟨n_M⟩⟨n_N⟩ in the main text, not only in Appendix C.
  4. [§III.C] The phrase 'no-feedback reference' is misleading because the reference is not the open-loop two-node system but an isolated birth–death process. Consider renaming it 'isolated-node reference' to avoid conflating direct regulation with absence of feedback, especially once the temporal comparison is revised.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central LNA noise decomposition is self-contained algebra; the temporal-baseline issue is a control concern, not a circular reduction.

full rationale

The central derivation is self-contained. Starting from the Langevin model (Eq. 1), the LNA covariance is obtained by solving the Lyapunov equation (Eq. 2; Appendix A). Eqs. (3)-(6) are an algebraic rearrangement of the resulting variance expressions: the intrinsic term is the Poisson baseline, the extrinsic term is the direct-regulation contribution, and the cyclic term T_MN H/(1-H) eta_int,N^2 is the remaining closed-loop correction. Nothing is fitted to the data appearing in Figs. 2-4; the Gillespie simulations are used only to test the LNA formulas, so the simulation agreement is an internal consistency check, not a fitted input called a prediction. The self-citations [29,30] are used for standard definitions of intrinsic/extrinsic noise and are not load-bearing; no uniqueness theorem or ansatz is imported from the authors' prior work. The temporal-signature argument (Sec. III.C, Eqs. 10-11) is weaker: the 'no-feedback' baseline rho^(0)=1-beta_N tau omits direct regulatory input, so the deviation captured by rho_hat includes both extrinsic propagation and cyclic noise, as the paper itself states in Eq. (11). This is a control/comparison concern about the temporal inference, not a circular reduction of the central claim to its inputs. The stated LNA limitation in the Conclusion is an acknowledged scope restriction and does not create circularity.

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

The framework is a pure analytical derivation from the LNA with no fitted parameters and no new physical entities. The 'cyclic noise' is a derived algebraic component of the covariance, not an independently postulated quantity. Key assumptions are the small-noise approximation, independence of noise sources, Hill-coefficient-1 regulation, and the short-time expansion used for autocorrelations.

assumptions (6)
  • domain assumption Linear noise approximation (LNA) holds: fluctuations are small and dynamics linearize around steady state
    Used to derive the Lyapunov equation (Eq. 2) and all covariance expressions; limits validity to small noise.
  • domain assumption Independent, temporally uncorrelated Gaussian noise sources with diagonal diffusion matrix D
    Noise processes for X and Y are assumed independent; D=diag(D_X,D_Y) is used in Eq. (2).
  • domain assumption Regulatory functions are Hill-type with Hill coefficient 1
    Table I specifies f_X and f_Y; this sets the form of sensitivities f′ but is not needed for the general decomposition structure.
  • standard math Steady-state balance f_N(⟨n⟩)=β_N⟨n_N⟩
    Used to write D_N=β_N⟨n_N⟩; standard LNA steady-state condition.
  • domain assumption Monostable regime with H<1 for reinforcing feedback
    Finite variances require det(J)>0; for H≥1 the Jacobian becomes singular (Section III.A).
  • domain assumption Short-time expansion e^{Jτ}≈I+Jτ with |λ_i|τ≪1
    Appendix C derives C(τ)≈(I+Jτ)Σ; the temporal-signature results are only asserted in this regime.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Feedback-mediated circulation and persistence of stochastic fluctuations in gene regulatory circuits." pith.science (2026). https://pith.science/paper/XOP66VTF

@misc{pith2026260717743,
  author       = {Pith},
  title        = {Pith review of: Feedback-mediated circulation and persistence of stochastic fluctuations in gene regulatory circuits},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XOP66VTF}},
  note         = {Machine review of arXiv:2607.17743}
}
read the original abstract

Feedback plays a significant role in biochemical networks that govern a multitude of cellular functions, including development, adaptation, and homeostasis. Yet, how feedback topology controls stochastic fluctuations remains incompletely understood. Here, we develop a theoretical framework for two-node feedback motifs composed of activating and repressive regulatory interactions between two transcription factors. Under the linear noise approximation, we identify a feedback-driven contribution to node-wise fluctuations, termed cyclic noise, that arises specifically from loop closure. Cyclic noise is the component of fluctuations that circulates through the regulatory circuit. Its sign and magnitude distinguish whether feedback amplifies or attenuates node-wise fluctuations. We further show that feedback-mediated noise circulation leaves a temporal signature in the decay of steady-state autocorrelation, revealing how loop closure modifies the persistence of fluctuations. We thus provide a minimal framework for understanding how feedback architecture regulates both the magnitude and the temporal persistence of noise in gene regulatory circuits.

Figures

Figures reproduced from arXiv: 2607.17743 by the authors.

Figure 1
Figure 1. FIG. 1. Schematic representation of two-node feedback motifs and feedback-dependent noise decomposition. (a) The four [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Dependence of cyclic noise on feedback-strength [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Dependence of steady-state autocorrelation on feedback strength [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Dependence of steady-state autocorrelation on timescale asymmetry [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references

  1. [1]

    M. B. Elowitz, A. J. Levine, E. D. Siggia, and P. S. Swain, Stochastic gene expression in a single cell, Science297, 1183 (2002)

    2002

  2. [2]

    Alon,An Introduction to Systems Biology: Design Principles of Biological Circuits(CRC Press, Boca Ra- ton, FL, 2006)

    U. Alon,An Introduction to Systems Biology: Design Principles of Biological Circuits(CRC Press, Boca Ra- ton, FL, 2006)

    2006

  3. [3]

    Becskei and L

    A. Becskei and L. Serrano, Engineering stability in gene networks by autoregulation, Nature405, 590 (2000)

    2000

  4. [4]

    Angeli, J

    D. Angeli, J. E. Ferrell, and E. D. Sontag, Detection of multistability, bifurcations, and hysteresis in a large class of biological positive-feedback systems, Proc. Natl. Acad. Sci. U.S.A.101, 1822 (2004)

    2004

  5. [5]

    J. L. Cherry and F. R. Adler, How to make a biological switch, J. Theor. Biol.203, 117 (2000)

    2000

  6. [6]

    Lestas, G

    I. Lestas, G. Vinnicombe, and J. Paulsson, Fundamen- tal limits on the suppression of molecular fluctuations, Nature467, 174 (2010)

    2010

  7. [7]

    A. Rhee, R. Cheong, and A. Levchenko, Noise decomposi- tion of intracellular biochemical signaling networks using nonequivalent reporters, Proc. Natl. Acad. Sci. U.S.A. 111, 17330 (2014)

    2014

  8. [8]

    Tkaˇ cik, A

    G. Tkaˇ cik, A. M. Walczak, and W. Bialek, Optimizing information flow in small genetic networks, Phys. Rev. E 80, 031920 (2009)

    2009

Show all 38 references
  1. [9]

    Smolen, D

    P. Smolen, D. A. Baxter, and J. H. Byrne, Interlinked dual-time feedback loops can enhance robustness to stochasticity and persistence of memory, Phys. Rev. E 79, 031902 (2009)

    2009

  2. [10]

    Chang, S

    D.-E. Chang, S. Leung, M. R. Atkinson, A. Reifler, D. Forger, and A. J. Ninfa, Building biological memory by linking positive feedback loops, Proc. Natl. Acad. Sci. U.S.A.107, 175 (2010)

    2010

  3. [11]

    Hinczewski and D

    M. Hinczewski and D. Thirumalai, Noise control in gene regulatory networks with negative feedback, J. Phys. Chem. B120, 6166 (2016)

    2016

  4. [12]

    C. Jia, P. Xie, M. Chen, and M. Q. Zhang, Stochastic fluctuations can reveal the feedback signs of gene regu- latory networks at the single-molecule level, Sci. Rep.7, 16037 (2017)

    2017

  5. [13]

    Cosentino and D

    C. Cosentino and D. Bates,Feedback control in systems 12 biology(CRC Press, 2011)

    2011

  6. [14]

    L. S. Tsimring, Noise in biology, Rep. Prog. Phys.77, 026601 (2014)

    2014

  7. [15]

    P. S. Swain, M. B. Elowitz, and E. D. Siggia, Intrinsic and extrinsic contributions to stochasticity in gene ex- pression, Proc. Natl. Acad. Sci. U.S.A.99, 12795 (2002)

    2002

  8. [16]

    T. J. Kobayashi, R. Yokota, and K. Aihara, Feedback regulation and its efficiency in biochemical networks, J. Stat. Phys.162, 1425 (2016)

    2016

  9. [17]

    Bintu, N

    L. Bintu, N. E. Buchler, H. G. Garcia, U. Gerland, T. Hwa, J. Kondev, and R. Phillips, Transcriptional reg- ulation by the numbers: models, Curr. Opin. Genet. Dev. 15, 116 (2005)

    2005

  10. [18]

    E. Ziv, I. Nemenman, and C. H. Wiggins, Optimal sig- nal processing in small stochastic biochemical networks, PLoS ONE2, e1077 (2007)

    2007

  11. [19]

    Tkaˇ cik, C

    G. Tkaˇ cik, C. G. Callan, and W. Bialek, Information flow and optimization in transcriptional regulation, Proc. Natl. Acad. Sci. U.S.A.105, 12265 (2008)

    2008

  12. [20]

    Tkaˇ cik, T

    G. Tkaˇ cik, T. Gregor, and W. Bialek, The role of input noise in transcriptional regulation, PLoS ONE3, e2774 (2008)

    2008

  13. [21]

    W. H. de Ronde, F. Tostevin, and P. R. Ten Wolde, Feed- forward loops and diamond motifs lead to tunable trans- mission of information in the frequency domain, Phys. Rev. E86, 021913 (2012)

    2012

  14. [22]

    A. M. Walczak, A. Mugler, and C. H. Wiggins, An- alytic methods for modeling stochastic regulatory net- works, Methods Mol. Biol.880, 273 (2012)

    2012

  15. [23]

    N. G. van Kampen,Stochastic Processes in Physics and Chemistry, 3rd ed.(North-Holland, Amsterdam, 2007)

    2007

  16. [24]

    Elf and M

    J. Elf and M. Ehrenberg, Fast evaluation of fluctuations in biochemical networks with the linear noise approxima- tion, Genome Res.13, 2475 (2003)

    2003

  17. [25]

    Paulsson, Summing up the noise in gene networks, Nature427, 415 (2004)

    J. Paulsson, Summing up the noise in gene networks, Nature427, 415 (2004)

    2004

  18. [26]

    P. S. Swain, Efficient attenuation of stochasticity in gene expression through post-transcriptional control, J. Mol. Biol.344, 965 (2004)

    2004

  19. [27]

    T˘ anase-Nicola, P

    S. T˘ anase-Nicola, P. B. Warren, and P. R. ten Wolde, Sig- nal detection, modularity, and the correlation between extrinsic and intrinsic noise in biochemical networks, Phys. Rev. Lett.97, 068102 (2006)

    2006

  20. [28]

    W. H. de Ronde, F. Tostevin, and P. R. ten Wolde, Effect of feedback on the fidelity of information transmission of time-varying signals, Phys. Rev. E82, 031914 (2010)

    2010

  21. [29]

    Nandi, S

    M. Nandi, S. Chattopadhyay, S. Bandyopadhyay, and S. K. Banik, Channel assisted noise propagation in a two- step cascade, Chaos34, 083128 (2024)

    2024

  22. [30]

    Nandi, Emergence of temporal noise hierarchy in co- regulated genes of multi-output feed-forward loop, Phys

    M. Nandi, Emergence of temporal noise hierarchy in co- regulated genes of multi-output feed-forward loop, Phys. Biol.22, 016006 (2024)

    2024

  23. [31]

    D. T. Gillespie, A general method for numerically simu- lating the stochastic time evolution of coupled chemical reactions, J. Comp. Phys.22, 403 (1976)

    1976

  24. [32]

    D. T. Gillespie, Exact stochastic simulation of coupled chemical reactions, J. Phys. Chem.81, 2340 (1977)

    1977

  25. [33]

    Alon, Network motifs: theory and experimental ap- proaches, Nat

    U. Alon, Network motifs: theory and experimental ap- proaches, Nat. Rev. Genet.8, 450 (2007)

    2007

  26. [34]

    Losick and C

    R. Losick and C. Desplan, Stochasticity and cell fate, Science320, 65 (2008)

    2008

  27. [35]

    Eldar and M

    A. Eldar and M. B. Elowitz, Functional roles for noise in genetic circuits, Nature467, 167 (2010)

    2010

  28. [36]

    T. S. Gardner, C. R. Cantor, and J. J. Collins, Construc- tion of a genetic toggle switch in Escherichia coli, Nature 403, 339 (2000)

    2000

  29. [37]

    D. T. Gillespie, The chemical langevin equation, J. Chem. Phys.113, 297 (2000)

    2000

  30. [38]

    Erban and S

    R. Erban and S. J. Chapman,Stochastic Modelling of Re- action–Diffusion Processes(Cambridge University Press,

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.