Support for viral persistence in bats from age-specific serology and models of maternal immunity

Spatiotemporally-localised prediction of virus emergence from wildlife requires focused studies on the ecology and immunology of reservoir hosts in their native habitat. Reliable predictions from mathematical models remain difficult in most systems due to a dearth of appropriate empirical data. Our goal was to study the circulation and immune dynamics of zoonotic viruses in bat populations and investigate the effects of maternally-derived and acquired immunity on viral persistence. Using rare age-specific serological data from wild-caught Eidolon helvum fruit bats as a case study, we estimated viral transmission parameters for a stochastic infection model. We estimated mean durations of around 6 months for maternally-derived immunity to Lagos bat virus and African henipavirus, whereas acquired immunity was long-lasting (Lagos bat virus: mean 12 years, henipavirus: mean 4 years). In the presence of a seasonal birth pulse, the effect of maternally-derived immunity on virus persistence within modelled bat populations was highly dependent on transmission characteristics. To explain previous reports of viral persistence within small natural and captive E. helvum populations, we hypothesise that some bats must experience prolonged infectious periods or within-host latency. By further elucidating plausible mechanisms of virus persistence in bat populations, we contribute to guidance of future field studies.


Supplementary Text 1:
MatAb have been directly or indirectly detected against henipaviruses in captive E.

Supplementary
To assess whether the mixed effects model is better than an ordinary regression model, we refit the latter using the gls function without the random intercept. The anova function was then be used to compare Akaike's information criterion (AIC) values. 2-6. Find the optimal random structure To assess whether the mixed effects model is better than an ordinary regression model, we refit the latter using the gls function without the random intercept. The anova function was then be used to compare Akaike's information criterion (AIC) values.

Check singularity
The definition of singularity is that some of the constrained parameters of the random effects theta parameters are on the boundary (equal to zero, or very very close to zero, say <10^-6): tt = getME(M2,'theta') ll = getME(M2,'lower') min(tt[ll==0])

Restart
Try restarting from previous fit, with maximum number of iterations.

Try a different optimizer
Try bobyqa for both phases -current GLMM default is bobyqa for first phase, Nelder-Mead for second phase.
M3 <-update(M2,start=ss,control=glmerControl(optimizer="bobyqa",optCtrl=list(maxf un=2e5))) summary ( The Random effects and Std Dev columns provide a measure of how much variability in the log(MFI) is due to Year and Country (the two random effects). Country has more variability than Year.
Best random effects structure is + (1|Country.f) 7-8. Find the optimal fixed structure Try removing Reproductive status as a fixed effect:  Model 8 (including Sex as a fixed effect) has a slightly lower AIC than with Sex removed, but this is not significant. Model 9 is the simpler model though.  No juveniles were sampled. Although sample sizes are small, age-class seroprevalence in Annobón is further supportive of active, endemic henipavirus transmission within sexually immature individuals. Insufficient seropositive individuals were detected to assess age-class seroprevalence for LBV on Annobón.

Visualisation of bootstrap estimates
All.mle.bs <-bind_rows( HNV.mle.bs %>% mutate(Virus="HNV"), LBV.mle.bs %>% mutate(Virus="LBV") ) All.mle.bs %>% ggpairs(mapping = aes(colour=Virus), columns=c(1:5),lower=list(continuous="densit y")) + ggtitle(" Supplementary Figure 4 proportion of acquired population immunity (PI) and infectious period (in days, IP) on the population size for which successful of invasion and persistence of infection is more probable than not (critical community size, CCS). Grey dotted lines show the mean duration of maternal immunity, as calculated in the age-specific immunity model (henipaviruses = 6.7 months). For some sets of parameter values, proportion of acquire population immunity (PI) and infectious period (in days, IP.d) on persistence of infection at a population size of 2500 individuals (as estimated for Eidolon helvum on Annobón island, Equatorial Guinea). Probability of pathogen extinction within 10 years of introduction (conditional on successful introduction) as a function of population size according to the colour scale shown. The probability of successful invasion decreased with increasing population immunity, meaning that for very high values of population immunity, none of the 1000 simulations were able to invade for some parameter values (grey areas in figure).
In B), following introduction in a naive population (PI = 0) comprising 2500 individuals or less, persistence is likely (>50% probability) for viruses with infectious period ≥40 days (IP.d = 40), regardless of maternal antibody (MatAb) protection. If the duration of MatAb is ≥6 months, then an infectious period of ≥50 days is likely required for persistence. In a non-naïve population with intermediate seroprevalences (e.g. PI = 0.4 or 0.5), extinction is only likely for infectious periods ≤ 10days.