arxiv: 2604.25714 · v2 · submitted 2026-04-28 · 🧬 q-bio.PE

Recognition: unknown

A modelling perspective on mosquito infectiousness: time-varying transmission competence in arbovirus vector

L\'ea Loisel , Tristan Monrocq , Vincent Raquin , Pauline Ezanno , Ga\"el Beaun\'ee

Authors on Pith no claims yet

Pith reviewed 2026-05-07 14:03 UTC · model grok-4.3

classification 🧬 q-bio.PE

keywords mosquito vector competencearbovirus transmissiontime-varying infectiousnessstochastic intra-vector modelchikungunya dengue zikatransmission cessation

0 comments

The pith

Models allowing mosquitoes to exit the transmitting state fit arbovirus vector competence data better than lifelong infectiousness in most declining cases.

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

The paper tests whether the standard assumption that mosquitoes stay infectious for life once virus reaches their saliva holds up against experimental data. It extends an existing stochastic model of virus spread inside the mosquito to allow either permanent loss or temporary interruption of the ability to transmit. When these versions are fitted to 52 experimental conditions across several viruses, the versions with an exit from the transmitting state match the observed late drops in transmitter proportions more closely in seven of the ten cases that show such drops. In those cases the estimated share of mosquitoes that successfully cross all internal barriers rises, while the time spent in earlier infected and disseminated states changes little. If the pattern reflects real biology, current models of disease spread may be under-counting the fraction of mosquitoes that ever become competent vectors.

Core claim

The authors introduce transmission states that permit either permanent cessation or temporary interruption of transmission into a stochastic intra-vector viral dynamics model. When the three competing models are fitted by Approximate Bayesian Computation to data from 52 vector competence conditions, models allowing exit from the transmitting state outperform the irreversible model in seven of the ten conditions that exhibit late declines in transmitter proportions, with clear improvement in five. In the selected cases posterior estimates of the proportion of mosquitoes that cross all barriers increase, while estimates of infected and disseminated state durations remain largely unchanged; non

What carries the argument

Addition of exit transitions from the transmitting state (permanent cessation or temporary interruption) inside the stochastic intra-vector viral dynamics model.

If this is right

In the cases where exit is favored, the estimated fraction of mosquitoes that become fully competent for transmission rises.
Durations of the infected and disseminated states stay similar across models.
Temporary interruption produces fits comparable to permanent cessation, with non-transmitting periods lasting several days.
The lifelong infectiousness assumption can underestimate the total number of mosquitoes that ever reach transmission competence.

Where Pith is reading between the lines

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

Epidemiological models that keep mosquitoes infectious for life may therefore underestimate the force of infection for some virus-mosquito combinations.
Targeted re-testing of the same mosquitoes at later times could distinguish permanent loss from temporary interruption.
If the loss is biological, it may be linked to mosquito immune clearance or viral dynamics inside the salivary glands, suggesting mechanistic follow-up experiments.

Load-bearing premise

The observed late-time declines in the proportion of transmitting mosquitoes are produced by a genuine biological loss of transmission competence rather than by experimental artifacts, mosquito mortality, or sampling variation.

What would settle it

Repeat the ten declining experiments with substantially larger late-time sample sizes, measure individual mosquito survival, and test whether the decline in transmitter proportion persists after correcting for mortality and sampling error.

Figures

Figures reproduced from arXiv: 2604.25714 by Ga\"el Beaun\'ee, L\'ea Loisel, Pauline Ezanno, Tristan Monrocq, Vincent Raquin.

**Figure 1.** Figure 1: Conceptual diagram of the intra-vector infection dynamic model view at source ↗

**Figure 2.** Figure 2: Comparison of visual fit between the three models (EIDT): permanent transmission, (EIDTP): permanent interruption of transmission, and view at source ↗

**Figure 3.** Figure 3: Posterior distributions of the barrier-crossing parameter across selected experimental conditions (scenarios) 1, 3, 4, 6 and 9 and model structures. view at source ↗

**Figure 4.** Figure 4: Distribution of transmitting-state duration across experimental view at source ↗

read the original abstract

Mosquito vector competence is usually represented as a process in which once virus is detected in saliva, mosquitoes are assumed to remain infectious for life, implying an irreversible transition to the transmitting state. However, some experiments report declines in the proportion of transmitting mosquitoes at late times post-exposure, suggesting transmission capacity may not be permanent. To investigate this hypothesis, we extended a previously developed stochastic intra-vector viral dynamics model by introducing transmission states allowing either permanent cessation or temporary interruption of transmission. We fitted three competing models to data from 52 vector competence conditions covering chikungunya, dengue, Zika, West Nile, and Rift Valley fever viruses, using Approximate Bayesian Computation with Sequential Monte Carlo inference. Among the 10 experimental conditions showing decline in transmitter proportions, models allowing exit from the transmitting state provided a better fit in 7 cases, with clear improvement in 5. In these cases, allowing interruption of transmission increased posterior estimates of the proportion of mosquitoes that crossed all intra-mosquito barriers, whereas estimates of infected and disseminated state durations were largely unchanged. In cases where intermittent transmission was selected, its performance was similar to that of permanent cessation with non-transmitting periods lasting several days. These results indicate that the assumption of lifelong mosquito infectiousness does not always provide the best explanation for vector competence data and may lead to underestimation of the proportion of mosquitoes that become capable of transmission. Incorporating time-varying transmission competence into intra-vector models could improve interpretation of vector competence experiments and refine epidemiological representations of arbovirus transmission.

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.

Desk Editor's Note private letter to a colleague

This paper adds exit states from the transmitting compartment and finds better fits in most declining datasets, but the declines may be sampling or mortality artifacts rather than biology.

read the letter

The main point is that extending the intra-vector model with reversible and irreversible exits from the transmitting state improves the fit in 7 of the 10 conditions that show late drops in transmitter proportions, and this raises the estimated share of mosquitoes that cross all barriers. They run ABC-SMC model comparison on 52 conditions spanning five arboviruses and show that intermittent or permanent cessation often beats the standard lifelong-infectiousness version, with similar performance between the two new variants and interruption durations of several days. The systematic check across many experiments is the useful part; it directly tests whether the usual assumption holds up in the data that modelers actually use. The increase in barrier-crossing estimates is a concrete change that would matter for downstream transmission models. The soft spot is the interpretation of those declines. Without per-time-point sample sizes or checks on whether a constant-rate model can produce apparent drops through binomial noise alone, it is hard to know if the exit parameters are capturing biology or just fitting noise. Mosquito mortality is not modeled separately, so any preferential loss of transmitters would be absorbed into the exit rate and inflate support for the more complex structures. The abstract also omits priors, tolerance schedules, and predictive checks, which leaves the robustness of the selection unclear even if the full methods section fills some gaps. This is for people who build or use arbovirus transmission models and need to decide how to represent vector competence. Readers working on dengue, Zika, or chikungunya forecasting would see the most direct value. It deserves a serious referee because it tests a common modeling choice against real experimental data with a clear comparison, even though the evidence for a biological exit needs more diagnostics on sample size and mortality before the conclusion can be taken as firm. I would send it to review but request raw counts, posterior predictive plots for the baseline model, and explicit handling of survival.

Referee Report

3 major / 1 minor

Summary. The paper extends a stochastic intra-vector viral dynamics model to include reversible or irreversible exit from the transmitting state. It applies ABC-SMC to compare three model structures (constant lifelong transmission, permanent cessation, temporary interruption) across 52 vector-competence conditions for chikungunya, dengue, Zika, West Nile, and Rift Valley fever viruses. In the 10 conditions exhibiting late-time declines in transmitter proportions, exit-allowed models are selected in 7 cases (clearly better in 5), yielding higher posterior estimates of the fraction of mosquitoes that cross all intra-mosquito barriers while leaving estimates of infected and disseminated durations largely unchanged.

Significance. If the central finding holds, the work provides a data-driven challenge to the standard assumption of permanent mosquito infectiousness for arboviruses. Systematic ABC-SMC comparison across 52 conditions spanning five viruses is a clear strength, offering reproducible model-selection evidence rather than reliance on isolated experiments. The result could improve both interpretation of vector-competence assays and epidemiological representations that currently treat transmission competence as irreversible.

major comments (3)

[Results (analysis of declining conditions)] Results, paragraph on the 10 declining conditions: The selection of exit-allowed models in 7 of 10 cases rests on observed late-time drops in transmitter proportions, yet the manuscript reports neither per-time-point sample sizes nor raw counts. Without these, it is impossible to determine whether the declines exceed binomial sampling variation expected under the constant-transmission null model.
[Methods (ABC-SMC inference)] Methods, ABC-SMC description: No prior distributions, tolerance schedules, or posterior-predictive diagnostics are provided for the model-comparison procedure. This omission directly affects assessment of whether the reported improvements in fit for the exit models are robust or sensitive to implementation choices.
[Model formulation] Model formulation section: The stochastic model does not appear to track survival of transmitting mosquitoes separately from non-transmitters. Consequently, any differential mortality that removes transmitters preferentially would be absorbed into the estimated exit rates, inflating support for the more complex models relative to the constant-transmission baseline.

minor comments (1)

[Abstract] The abstract states that 'allowing interruption of transmission increased posterior estimates' but does not quantify the magnitude of the increase or provide a table of median posterior shifts across the five clear-improvement cases.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive and detailed comments. These have prompted us to improve the documentation of our methods and data, and to clarify model assumptions. We respond to each major comment below.

read point-by-point responses

Referee: Results (analysis of declining conditions): The selection of exit-allowed models in 7 of 10 cases rests on observed late-time drops in transmitter proportions, yet the manuscript reports neither per-time-point sample sizes nor raw counts. Without these, it is impossible to determine whether the declines exceed binomial sampling variation expected under the constant-transmission null model.

Authors: We agree that per-time-point sample sizes should be reported for transparency. The data originate from published studies, and we will add a supplementary table in the revised manuscript that lists the number of mosquitoes assayed at each time point for all 10 declining conditions (sourced from the original references). The ABC-SMC procedure already incorporates stochastic sampling variation when simulating data under each model and computing posterior model probabilities; model selection is therefore not based solely on visual late-time declines but on the full time-series fit. We have revised the Results section to reference this table and note that the observed late-time proportions in the selected cases fall outside the range typically produced by the constant-transmission model under binomial sampling. revision: yes
Referee: Methods (ABC-SMC inference): No prior distributions, tolerance schedules, or posterior-predictive diagnostics are provided for the model-comparison procedure. This omission directly affects assessment of whether the reported improvements in fit for the exit models are robust or sensitive to implementation choices.

Authors: We acknowledge the omission of these implementation details. In the revised Methods section we will add a new subsection that specifies (i) the prior distributions for all parameters (chosen from biological ranges reported in the literature), (ii) the tolerance schedule and number of particles used in the ABC-SMC runs, and (iii) the results of posterior-predictive checks performed on the retained particles for each model. These additions will allow readers to evaluate the robustness of the model-selection outcomes. revision: yes
Referee: Model formulation: The stochastic model does not appear to track survival of transmitting mosquitoes separately from non-transmitters. Consequently, any differential mortality that removes transmitters preferentially would be absorbed into the estimated exit rates, inflating support for the more complex models relative to the constant-transmission baseline.

Authors: This is a legitimate identifiability concern. Our model assumes a constant per-capita mortality rate that does not depend on transmission state, which is the standard simplification in intra-vector dynamics models. Differential mortality would indeed be confounded with the exit rates we estimate. However, because the constant-transmission model forbids any exit (including mortality-driven loss of transmitters), any such differential effect would systematically worsen its fit relative to the exit-allowed models; the model comparison therefore remains informative even under this limitation. We will revise the Model formulation section to state the mortality assumption explicitly and add a limitations paragraph discussing possible confounding with differential survival, recommending collection of state-specific survival data in future experiments. revision: partial

Circularity Check

0 steps flagged

No significant circularity detected in model extension or selection

full rationale

The paper extends a cited prior stochastic intra-vector model by adding new transmission states (permanent cessation or temporary interruption) and then applies ABC-SMC to fit three distinct model structures to the 52 experimental time-series conditions. Model selection among the 10 declining cases is performed by direct comparison of posterior fit quality on the observed proportions; this is a standard data-driven procedure that does not reduce any reported result to a definitional equivalence or to a fitted parameter renamed as a prediction. No equations are shown to be tautological, no uniqueness theorem is imported from self-citation, and the central claim (better fit for exit-allowed models in 7/10 cases) remains an independent inference step rather than a self-referential loop. The prior model citation supplies the base dynamics but is not load-bearing for the new comparison.

Axiom & Free-Parameter Ledger

3 free parameters · 2 axioms · 2 invented entities

The central claim rests on standard stochastic compartmental assumptions plus two new transmission-exit states whose parameters are estimated from the same data used for model comparison.

free parameters (3)

duration of transmitting state
Fitted parameter governing how long mosquitoes remain able to transmit before possible exit.
probability of permanent cessation
Fitted parameter controlling the rate of irreversible loss of competence.
duration of non-transmitting interruption
Fitted parameter for the length of temporary pauses in transmission.

axioms (2)

domain assumption Experimental declines in transmitter proportions reflect true loss of competence rather than measurement error or host death.
Invoked when interpreting late-time data as evidence for exit states.
standard math The intra-vector viral dynamics can be represented by a small number of discrete states with exponential waiting times.
Standard assumption of the base stochastic model being extended.

invented entities (2)

reversible non-transmitting state no independent evidence
purpose: To allow temporary interruption of transmission after the mosquito has become competent.
New compartment added to explain intermittent loss of saliva positivity.
irreversible non-transmitting state no independent evidence
purpose: To allow permanent cessation of transmission.
New absorbing state added to capture lifelong loss of competence.

pith-pipeline@v0.9.0 · 5590 in / 1630 out tokens · 78766 ms · 2026-05-07T14:03:48.499788+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

22 extracted references

[1]

Amraoui, F., Atyame-Nten, C., Vega-Rúa, A., Lourenço-de Oliveira, R., Vazeille, M., and Failloux, A. B. (2016). Culex mosquitoes are experimen- tally unable to transmit Zika virus.Euro Surveillance: Bulletin Europeen Sur Les Maladies Transmissibles = European Communicable Disease Bulletin, 21(35):30333. Amraoui, F., Ben Ayed, W., Madec, Y., Faraj, C., Him...

2016
[2]

h., Sene, N

Diallo, D., Diouf, B., Gaye, A., NDiaye, E. h., Sene, N. M., Dia, I., and Diallo, M. (2022). Dengue vectors in Africa: A review.Heliyon, 8(5):e09459. Eastwood, G., Kramer, L. D., Goodman, S. J., and Cunningham, A. A. (2011). West Nile Virus Vector Competency of Culex quinquefasciatus Mosquitoes in the Galápagos Islands.The American Journal of Tropical Med...

2022
[3]

Fontaine, A., Lequime, S., Moltini-Conclois, I., Jiolle, D., Leparc-Goffart, I., Jr, R. C. R., and Lambrechts, L. (2018). Epidemiological significance of dengue virus genetic variation in mosquito infection dynamics.PLOS Pathogens, 14(7):e1007187. lecture totale annotée. Fortuna, C., Severini, F., Marsili, G., Toma, L., Amendola, A., Venturi, G., Argentin...

2018
[4]

Gubler, D. J. (2002). The Global Emergence/Resurgence of Arboviral Diseases As Public Health Problems.Archives of Medical Research, 33(4):330–342. Hall-Mendelin, S., Pyke, A. T., Moore, P. R., Mackay, I. M., McMahon, J. L., Ritchie, S. A., Taylor, C. T., Moore, F. A., and van den Hurk, A. F. (2016). Assessment of Local Mosquito Species Incriminates Aedes ...

2002
[5]

P., Skinner, E

Kain, M. P., Skinner, E. B., Athni, T. S., Ramirez, A. L., Mordecai, E. A., and van den Hurk, A. F. (2022). Not all mosquitoes are created equal: A synthesis of vector competence experiments reinforces virus associations of Australian mosquitoes.PLoS neglected tropical diseases, 16(10):e0010768. Kramer, L. D. and Ebel, G. D. (2003). Dynamics of Flavivirus...

2022
[6]

Wang, Y., Ling, L., Zhang, Z., and Marin-Lopez, A. (2022). Current Ad- vances in Zika Vaccine Development.V accines, 10(11). Wu, P., Yu, X., Wang, P., and Cheng, G. (2019). Arbovirus lifecycle in mosquito: Acquisition, propagation and transmission.Expert Reviews in Molecular Medicine,

2022
[7]

Y., Chen, B., Christofferson, R., Ebel, G., Fagre, A

Wu, V. Y., Chen, B., Christofferson, R., Ebel, G., Fagre, A. C., Gallichotte, E. N., Sweeny, A. R., Carlson, C. J., and Ryan, S. J. (2022). A minimum data standard for vector competence experiments.Scientific Data, 9(1):634. Number:

2022
[8]

Zakhia, R., Mousson, L., Vazeille, M., Haddad, N., and Failloux, A.-B. (2018). Experimental transmission of West Nile Virus and Rift Valley Fever Virus by Culex pipiens from Lebanon.PLoS neglected tropical diseases, 12(1):e0005983. 12 Léa Loiselet al. Table S1.Summary of experimental data used to infer the three IVD model for chikungunya virus (CHIKV). ID...

2018
[9]

(British Virgin Islands) (Mexico, Poza Rica) 70-80 D:legs/wings(RT-PCR) 12h:12h T:saliva (PFA) 2 CHIKV CarIls_2013_aeg_Guy_6 Carribean strain-2013 Ae.aegypti 6 28°c 3,5,7,10,14 41 I:body(PFA) (Wang et al.,

2013
[10]

(Reunion Island) (Tunisia,Car, Ami, La Mar) 80 D:thorax/head (FFA) 16h:8h T:saliva (FFA) 8 CHIKV Ncal_aeg_Mad_Fun_7.3 NC/2011-568 Ae.aegypti 7.3 28 +/- 1°c 3,6,9,14 20 I:thorax/abdomen (ND) (Seixas et al.,

2011
[11]

(New Caledonia) (Mad Island, Funchal) 80 I:head(FFA) ND T:saliva(FFA) 9 CHIKV Ncal_aeg_Mad_Pa_do_Ma_7.3 NC/2011-568 Ae.aegypti 7.3 28 +/- 1°c 3,6,9,14 20 I:thorax/abdomen (ND) (Seixas et al.,

2011
[12]

(Tunisia,Car, Ami, La Mar) 80 D:thorax/head (FFA) 16h:8h T:saliva (FFA) 16 DENV-3 Cair_1998_aeg_Cair_4.7 1998- GenBank JN406514 Ae aegypti 4.7 28°c 2,3,4,5,6,7,10,14 21 I:body(ELISA) (Ritchie et al.,

1998
[13]

(Egypt) (France) 80% D:head PFA) 12h:12h T:saliva (PFA) Modelling intermittent mosquito infectiousness15 Table S4.Summary of experimental data used to infer the three IVD model for West-Nile virus (WNV). ID:Infectious Dose (Log10 FFU/mL), FFA : Fluorescent Focus Assay, RTPCR:reverse transcription polymerase chain reaction Dpe: Day post exposition, ND:not ...

1956
[14]

(USA) Galapagos D: legs(PFA) T : saliva(PFA) 32 WNV WN0195_CulPip_SanCru_7.65_30_A 02-1956 Culex pipiens quinquefasciatus 7,65 30°C 7,11,14,21 40 I : abdomen (PFA) (Eastwood et al.,

1956
[15]

(West America) (USA) D: legs(PFA) T : saliva(PFA) 37 WNV WNVWN0_CulTar_CalUSA_7_27_A 02-1986 , 02-1956, Am Culex tarsalis 7 27 5,7,9,14 185 I : abdomen (PFA) (Moudy et al.,

1986
[16]

Table S5.Summary of experimental data used to infer the three IVD model for Zika virus (ZIKV)

(North America) (USA) D: legs(PFA) T : saliva(PFA) 16 Léa Loiselet al. Table S5.Summary of experimental data used to infer the three IVD model for Zika virus (ZIKV). ID:Infectious Dose (Log10 FFU/mL), FFA : Fluorescent Focus Assay, RTPCR:reverse transcription polymerase chain reaction Dpe: Day post exposition, ND:not defined Experimental conditionsVirus s...

2016
[17]

(New Caledonia) 80 D:head PFA) 12h:12h T:saliva (PFA) 41 ZIKV Ncal_2015_aeg_Ncal_7 NC-2015-2391 Ae.aegypti 7 28°c 6,9,14,21 34 I:abdomen/thorax (PFA) (Calvez et al., 2018b) (New Caledonia,

2015
[18]

(New Caledonia) 80 D:head PFA) 12h:12h T:saliva (PFA) 42 ZIKV Ncal_aeg_FrP_7 NC-2014-5132 Ae.aegypti 7 28°c 6,9,14,21 32 I:abdomen/thorax (PFA) (Calvez et al., 2018a) (New Caledonia) (French Polynesia) 80 D:head PFA) 12h:12h T:saliva (PFA) 43 ZIKV Ncal_aeg_Ncal_7 NC-2014-5132 Ae.aegypti 7 28°c 6,9,14,21 26 I:abdomen/thorax (PFA) (Calvez et al., 2018a) (Ne...

2014
[19]

(New Caledonia) (Morocco, Rabat) 80 D:head (PFA) 16h:8h T:saliva (PFA) 46 ZIKV Ncal_albo_Tun_7 NC-2014-5132 Ae.albopictus 7 28 +/- 1°c 7,10,14,21 23 I:abdomen (PFA) (Bohers et al.,

2014
[20]

(New Caledonia) (Tunisia,Car, Ami, La Mar) 80 D:thorax/head (PFA) 16h:8h T:saliva (PFA) 47 ZIKV Ncal_pol_FrP_7 NC-2014-5132 Ae.polynesiensis 7 28°c 6,9,14,21 30 I:abdomen/thorax (PFA) (Calvez et al., 2018a) (New Caledonia) (French Polynesia) 80 D:head PFA) 12h:12h T:saliva (PFA) 48 ZIKV Ncal_pol_Wal_7 NC-2014-5132 Ae.polynesiensis 7 28°c 6,9,14,21 38 I:ab...

2014
[21]

(United States Puerto Rico) (Mexico, Poza Rica) 70-80 D:legs/wings (RT-PCR) 12h:12h T:saliva (PFA) 50 ZIKV Sen_1991_aeg_Ncal_7 AF-1991-HD78788 Ae.aegypti 7 28°c 6,9,14,21 28 I:abdomen/thorax (PFA) (Calvez et al., 2018b) (Senegal,

1991
[22]

(New Caledonia) 80 D:head PFA) 12h:12h T:saliva (PFA) 51 ZIKV Sen_2002_aeg_Ncal_7 AF-2002-ArD 165 522 Ae.aegypti 7 28°c 6,9,14,21 30 I:abdomen/thorax (PFA) (Calvez et al., 2018b) (Senegal,

2002