ErrorMsg: running ART extension w/ nodefactor and nodematch

[64Zone]

Hi There,

I am trying to incorporate the HIV ART extension (https://github.com/EpiModel/EpiModel-Gallery/tree/master/2019-03-HIV) to a base EpiModel with a nodal attribute (i.e., high vs. low risk). I see the following error message when calling netsim():

Error in { :
task 1 failed - "In term 'nodefactor' in package 'ergm' (called from term 'Form (dynamic)', called from term 'Form' in package 'tergm'): Attribute '"risk"' has missing data, which is not currently supported by ergm."

I wonder if you can hint on what the error means? Or, is it possible to have the ART extension with nodal attributes? Here is how I initialize the conditions and my sessionInfo(). Thanks much!

---

h <- 600
k <- 100
status <- c(rep("i", k), rep("s", 2970-k), rep("i", h-k), rep("s", 2530-h+k))
init_ART <- init.net(status.vector = status)

control_ART <- control.net(type = NULL,
nsteps = 520,
nsims = 4,
ncores = 5,
infection.FUN = infect,
progress.FUN = progress,
departures.FUN = dfunc,
arrivals.FUN = afunc,
resimulate.network = TRUE,
verbose = TRUE,
epi.by = "risk",
module.order = c("resim_nets.FUN", "progress.FUN",
"infection.FUN", "departures.FUN",
"arrivals.FUN", "prevalence.FUN"))

param_ART <- param.net(inf.prob.chronic = 0.01,
relative.inf.prob.acute = 10,
relative.inf.prob.AIDS = 5,
relative.inf.prob.ART = 0.05,
act.rate = 4,
AcuteToChronic1.Rate = 1/12,
Chronic1ToChronic2.Rate = 1/260,
Chronic2ToAIDS.Rate = 1/260,
ART.Treatment.Rate = 0.01,
ART.Discontinuance.Rate = 0.005,
ART.Progression.Reduction.Rate = 0.5,
AIDSToDepart.Rate = 1/104,
arrival.rate = 0.002,
departure.rate = 0.003)

sim_ART <- netsim(est1_ART, param_ART, init_ART, control_ART)

R version 4.1.2 (2021-11-01)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 19043)

Matrix products: default

locale:
[1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 LC_MONETARY=English_United States.1252 LC_NUMERIC=C
[5] LC_TIME=English_United States.1252

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] EpiModel_2.2.1 statnet.common_4.5.0 tergm_4.0.2 ergm_4.1.2 networkDynamic_0.11.1 network_1.17.1 deSolve_1.31 forcats_0.5.1
[9] stringr_1.4.0 dplyr_1.0.7 purrr_0.3.4 readr_2.1.2 tidyr_1.2.0 tibble_3.1.6 ggplot2_3.3.5 tidyverse_1.3.1

loaded via a namespace (and not attached):
[1] httr_1.4.2 jsonlite_1.7.3 foreach_1.5.2 modelr_0.1.8 assertthat_0.2.1 cellranger_1.1.0 yaml_2.2.2
[8] robustbase_0.93-9 pillar_1.7.0 backports_1.4.1 lattice_0.20-45 glue_1.6.1 digest_0.6.29 RColorBrewer_1.1-2
[15] rvest_1.0.2 colorspace_2.0-2 htmltools_0.5.2 Matrix_1.3-4 pkgconfig_2.0.3 broom_0.7.12 haven_2.4.3
[22] scales_1.1.1 tzdb_0.2.0 generics_0.1.2 ellipsis_0.3.2 cachem_1.0.6 withr_2.4.3 lazyeval_0.2.2
[29] cli_3.1.1 magrittr_2.0.2 crayon_1.4.2 readxl_1.3.1 evaluate_0.14 memoise_2.0.1 fs_1.5.2
[36] fansi_1.0.2 doParallel_1.0.17 nlme_3.1-153 MASS_7.3-54 xml2_1.3.3 tools_4.1.2 hms_1.1.1
[43] lifecycle_1.0.1 trust_0.1-8 munsell_0.5.0 reprex_2.0.1 compiler_4.1.2 rlang_1.0.1 grid_4.1.2
[50] rstudioapi_0.13 iterators_1.0.14 rmarkdown_2.11 gtable_0.3.0 codetools_0.2-18 DBI_1.1.2 R6_2.5.1
[57] lpSolveAPI_5.5.2.0-17.7 rle_0.9.2 lubridate_1.8.0 knitr_1.37 fastmap_1.1.0 utf8_1.2.2 ape_5.6-2
[64] stringi_1.7.6 parallel_4.1.2 Rcpp_1.0.8 vctrs_0.3.8 DEoptimR_1.0-10 dbplyr_2.1.1 tidyselect_1.1.1
[71] xfun_0.29 coda_0.19-4

[sgoodreau]

Hello,

I don't know all of the ins and outs of the current state of EpiModelHIV code, but I can parse the error you are seeing. It means that at least one node in the model does not have the attribute risk set for it. Is risk an attribute that you added to the model? If so, then you need to make sure that your code assigns it both to the nodes that exist at the start of the simulation, and also to the nodes that enter into the population at later timepoints. The latter would typically be done by modifying the function afunc to your own version (e.g. afunc.64zone with an added line assigning values of risk to the new arrivals, and then calling arrivals.FUN = afunc.64zone instead of arrivals.FUN = afunc.

[64Zone]

Thank you Steven. I think this is an insightful parsing! :)

Best,
Sicong