Spam.mpxv
Stochastic pair approximation model for Monkeypox virus
Install / Use
/learn @mrc-ide/Spam.mpxvREADME
spam.mpxv: Stochastic pair approximation model for Monkeypox virus
Population structure is modelled as a dynamic network based on behavioural surveillance of MSM. Each individual has a mixture of long-duration and short-duration(casual) contacts as well as one-off contacts.
Degree distributions and partnership durations are based on: Anderson et al., Epidemiology 2021
Relative contact rates for different contact types are based on Jenness et al., JID 2016
The model is designed within the ‘edge-based compartmental’(EBCM) framework laid out in
- Miller JC, Volz EM. Model hierarchies in edge-based compartmental modeling for infectious disease spread. J Math Biol. 2013;67: 869–899. doi:10.1007/s00285-012-0572-3
- Volz E, Meyers LA. Susceptible–infected–recovered epidemics in dynamic contact networks. Proceedings of the Royal Society B. 2007. Available: https://royalsocietypublishing.org/doi/abs/10.1098/rspb.2007.1159
A system of stochastic differential equations were designed around the EBCM by introducing Guassian noise on 1) the force of infection and 2) the average degree among new infected individuals.
The model is designed to work with the pomp package for simulation and
model fitting.
Example simulation
library( spam.mpxv )
set.seed( 1111 )
st0 <- Sys.time()
# Model m1.0 implements the first basic model with three contact types
# Default parameters are not calibrated and only for demonstration
s1 <- pomp::simulate( m3.0 )
st1 <- Sys.time()
print( st1 - st0 )
## Time difference of 0.01085997 secs
Reproduction number
calcR0.3.0(beta = 2.25, gamma1 = 1/4)
## [1] 1.564376
plot( s1 )
Example simulated data:
s1d <- as.data.frame( s1 )
print( tail( s1d ))
## time Ytravel Yendog Yunk thetaf MSEf MEf MSSf MSIf MIf thetag MSEg MEg
## 85 85 0 5 0 0.9998131 9.170162e-05 0.0001514393 0.9993615 1.511577e-05 6.324420e-05 0.9995270 0.0002672865 0.0002543847
## 86 86 0 5 0 0.9998024 9.173478e-05 0.0001544219 0.9993407 1.606797e-05 6.636306e-05 0.9995097 0.0002803447 0.0002666551
## 87 87 1 5 0 0.9997917 9.982899e-05 0.0001650752 0.9993119 1.635344e-05 6.907504e-05 0.9994924 0.0002969041 0.0002824125
## 88 88 0 5 0 0.9997757 8.973335e-05 0.0001581575 0.9992994 1.746761e-05 7.244068e-05 0.9994786 0.0002866366 0.0002712892
## 89 89 0 2 0 0.9997650 9.114120e-05 0.0001611069 0.9992778 1.653199e-05 7.410020e-05 0.9994579 0.0002924269 0.0002764542
## 90 90 0 2 0 0.9997650 9.570996e-05 0.0001671799 0.9992526 1.647858e-05 7.571351e-05 0.9994337 0.0002984528 0.0002818882
## MSSg MSIg MIg thetah MEh MIh E I newI cutf cutg cuth cuts seedrate
## 85 0.9987887 5.548255e-05 0.0001009370 0.9830379 0.0003860357 0.0001774621 72.12746 31.61915 9.015933 35 137 69 75 0.8608633
## 86 0.9987367 5.864850e-05 0.0001075009 0.9826876 0.0003658050 0.0001813510 73.11153 32.85330 9.138941 37 142 72 75 0.7536780
## 87 0.9986794 6.173658e-05 0.0001139575 0.9819873 0.0003957859 0.0001817389 76.97259 34.26155 9.621573 39 147 78 75 0.7162264
## 88 0.9986469 6.522938e-05 0.0001207697 0.9816373 0.0003871005 0.0001857774 77.35101 35.36504 9.668877 42 151 81 75 0.7354071
## 89 0.9985993 6.554290e-05 0.0001244884 0.9812875 0.0004145907 0.0001877206 78.68214 36.35905 9.835267 44 157 84 75 0.8342455
## 90 0.9985503 6.647204e-05 0.0001279231 0.9808213 0.0003547351 0.0001926143 79.84687 37.25014 9.980859 44 164 85 78 0.7713144
How long does it take to run a particle filter on the simulated data?
print(
system.time(pfilter(s1,Np=100))
)
## user system elapsed
## 0.884 0.000 0.879
Version 0.0.3 2022-06-20
- Revise natural history parameters, SEIR
