dualDiscrete: Dual-host pathogen in structured (discrete) hosts populations
In nosoi: A Forward Agent-Based Transmission Chain Simulator

Description Usage Arguments Details Value Structure Matrix Order of Arguments Structure Parameters Suffixes See Also Examples

View source: R/dual-discrete.R

This function, that can be wrapped within nosoiSim, runs a dual-host transmission chain simulation, with discrete hosts populations structures (e.g. spatial, socio-economic, etc.). The simulation stops either at the end of given time (specified by length.sim) or when the number of hosts infected threshold (max.infected) is crossed.

dualDiscrete(
  length.sim,
  max.infected.A,
  max.infected.B,
  init.individuals.A,
  init.individuals.B,
  init.structure.A,
  init.structure.B,
  structure.matrix.A,
  structure.matrix.B,
  pExit.A,
  param.pExit.A,
  timeDep.pExit.A = FALSE,
  diff.pExit.A = FALSE,
  hostCount.pExit.A = FALSE,
  pMove.A,
  param.pMove.A,
  timeDep.pMove.A = FALSE,
  diff.pMove.A = FALSE,
  hostCount.pMove.A = FALSE,
  nContact.A,
  param.nContact.A,
  timeDep.nContact.A = FALSE,
  diff.nContact.A = FALSE,
  hostCount.nContact.A = FALSE,
  pTrans.A,
  param.pTrans.A,
  timeDep.pTrans.A = FALSE,
  diff.pTrans.A = FALSE,
  hostCount.pTrans.A = FALSE,
  prefix.host.A = "H",
  pExit.B,
  param.pExit.B,
  timeDep.pExit.B = FALSE,
  diff.pExit.B = FALSE,
  hostCount.pExit.B = FALSE,
  pMove.B,
  param.pMove.B,
  timeDep.pMove.B = FALSE,
  diff.pMove.B = FALSE,
  hostCount.pMove.B = FALSE,
  nContact.B,
  param.nContact.B,
  timeDep.nContact.B = FALSE,
  diff.nContact.B = FALSE,
  hostCount.nContact.B = FALSE,
  pTrans.B,
  param.pTrans.B,
  timeDep.pTrans.B = FALSE,
  diff.pTrans.B = FALSE,
  hostCount.pTrans.B = FALSE,
  prefix.host.B = "V",
  print.progress = TRUE,
  print.step = 10
)

`length.sim`	specifies the length (in unit of time) over which the simulation should be run.
`max.infected.A`	specifies the maximum number of individual hosts A that can be infected in the simulation.
`max.infected.B`	specifies the maximum number of individual hosts B that can be infected in the simulation.
`init.individuals.A`	number of initially infected individuals (hosts A).
`init.individuals.B`	number of initially infected individuals (hosts B).
`init.structure.A`	in which state (e.g. location) the initially infected individuals of host-type A are located (`NA` if init.individual.A is 0)?
`init.structure.B`	in which state (e.g. location) the initially infected individuals of host-type B are located (`NA` if init.individual.B is 0)?
`structure.matrix.A`	transition matrix (probabilities) to go from location A (row) to B (column) for host-type A.
`structure.matrix.B`	transition matrix (probabilities) to go from location A (row) to B (column) for host-type B.
`pExit.A`	function that gives the probability to exit the simulation for an infected host A (either moving out, dying, etc.).
`param.pExit.A`	parameter names (list of functions) for the pExit for host-type A.
`timeDep.pExit.A`	is pExit of host-type A dependent on the absolute time of the simulation (TRUE/FALSE)?
`diff.pExit.A`	is pExit of host-type A different between states of the structured population (TRUE/FALSE)?
`hostCount.pExit.A`	does pExit of host-type A vary with the host count (of either host-type A or B) in the state? (TRUE/FALSE); diff.pExit.A should be TRUE.
`pMove.A`	function that gives the probability of a host moving as a function of time for host-type A.
`param.pMove.A`	parameter names (list of functions) for the pMove for host-type A.
`timeDep.pMove.A`	is pMove of host-type A dependent on the absolute time of the simulation (TRUE/FALSE)?
`diff.pMove.A`	is pMove of host-type A different between states of the structured population (TRUE/FALSE)?
`hostCount.pMove.A`	does pMove of host-type A vary with the host count (of either host A or B) in the state? (TRUE/FALSE); diff.pMove.A should be TRUE.
`nContact.A`	function that gives the number of potential transmission events per unit of time for host-type A.
`param.nContact.A`	parameter names (list of functions) for param.nContact for host-type A.
`timeDep.nContact.A`	is nContact of host-type A dependent on the absolute time of the simulation (TRUE/FALSE)?
`diff.nContact.A`	is nContact of host-type A different between states of the structured population (TRUE/FALSE)?
`hostCount.nContact.A`	does nContact of host-type A vary with the host count (of either host A or B) in the state? (TRUE/FALSE); diff.nContact.A should be TRUE.
`pTrans.A`	function that gives the probability of transmit a pathogen as a function of time since infection for host A.
`param.pTrans.A`	parameter names (list of functions) for the pExit for host A.
`timeDep.pTrans.A`	is pTrans of host-type A dependent on the absolute time of the simulation (TRUE/FALSE)?
`diff.pTrans.A`	is pTrans of host-type A different between states of the structured population (TRUE/FALSE)?
`hostCount.pTrans.A`	does pTrans of host-type A vary with the host count (of either host A or B) in the state? (TRUE/FALSE); diff.pTrans.A should be TRUE.
`prefix.host.A`	character(s) to be used as a prefix for the host A identification number.
`pExit.B`	function that gives the probability to exit the simulation for an infected host B (either moving out, dying, etc.).
`param.pExit.B`	parameter names (list of functions) for the pExit for host-type B.
`timeDep.pExit.B`	is pExit of host-type B dependent on the absolute time of the simulation (TRUE/FALSE)?
`diff.pExit.B`	is pExit of host-type B different between states of the structured population (TRUE/FALSE)?
`hostCount.pExit.B`	does pExit of host-type B vary with the host count (of either host A or B) in the state? (TRUE/FALSE); diff.pExit.B should be TRUE.
`pMove.B`	function that gives the probability of a host moving as a function of time for host-type B.
`param.pMove.B`	parameter names (list of functions) for the pMove for host-type B.
`timeDep.pMove.B`	is sdMove of host-type B dependent on the absolute time of the simulation (TRUE/FALSE) for host-type B.
`diff.pMove.B`	is pMove of host-type B different between states of the structured population (TRUE/FALSE)?
`hostCount.pMove.B`	does pMove of host-type B vary with the host count (of either host A or B) in the state? (TRUE/FALSE); diff.pMove.B should be TRUE.
`nContact.B`	function that gives the number of potential transmission events per unit of time for host B.
`param.nContact.B`	parameter names (list of functions) for param.nContact for host-type B.
`timeDep.nContact.B`	is nContact of host-type B dependent on the absolute time of the simulation (TRUE/FALSE)?
`diff.nContact.B`	is nContact of host-type B different between states of the structured population (TRUE/FALSE)?
`hostCount.nContact.B`	does nContact of host-type B vary with the host count (of either host A or B) in the state? (TRUE/FALSE); diff.nContact.B should be TRUE.
`pTrans.B`	function that gives the probability of transmit a pathogen as a function of time since infection for host B.
`param.pTrans.B`	parameter names (list of functions) for the pExit for host-type B.
`timeDep.pTrans.B`	is pTrans of host-type B dependent on the absolute time of the simulation (TRUE/FALSE)?
`diff.pTrans.B`	is pTrans host-type B different between states of the structured population (TRUE/FALSE)?
`hostCount.pTrans.B`	does pTrans of host-type B vary with the host count (of either host A or B) in the state? (TRUE/FALSE); diff.pTrans.B should be TRUE.
`prefix.host.B`	character(s) to be used as a prefix for the host B identification number.
`print.progress`	if TRUE, displays a progress bar (current time/length.sim).
`print.step`	print.progress is TRUE, step with which the progress message will be printed.

The pExit and pTrans functions should return a single probability (a number between 0 and 1), and nContact a positive natural number (positive integer) or 0.

The param arguments should be a list of functions or NA. Each item name in the parameter list should have the same name as the argument in the corresponding function.

The use of timeDep (switch to TRUE) makes the corresponding function use the argument prestime (for "present time").

An object of class nosoiSim, containing all results of the simulation.

The structure/transition matrix provided provided should of class matrix, with the same number of rows and columns, rows representing departure state and column the arrival state. All rows should add to 1. Probability values can be different for hosts A and B (so two different matrices), but the name of the column and the rows should be shared.

The user specified function's arguments should follow this order: t (mandatory), prestime (optional, only if timeDep is TRUE), current.in (optional, only if diff is TRUE), host.count.A or host.count.B (optional, only if hostCount is TRUE) and parameters specified in the list.

The pMove function should return a single probability (a number between 0 and 1).

The use of diff (switch to TRUE) makes the corresponding function use the argument current.in (for "currently in"). Your function should in that case give a result for every possible discrete state.

The use of hostCount (switch to TRUE) makes the corresponding function use the argument host.count.

The suffix .A or .B specifies if the considered function or parameter concerns host type A or B.

For simulations with a structure in continuous space, see dualContinuous. For simulations without any structures, see dualNone.

#Host A
t_infectA_fct <- function(x){rnorm(x,mean = 12,sd=3)}
pTrans_hostA <- function(t,t_infectA){
 if(t/t_infectA <= 1){p=sin(pi*t/t_infectA)}
 if(t/t_infectA > 1){p=0}
 return(p)
}

p_Move_fctA  <- function(t){return(0.1)}

p_Exit_fctA  <- function(t,t_infectA){
 if(t/t_infectA <= 1){p=0}
 if(t/t_infectA > 1){p=1}
 return(p)
}

time_contact_A = function(t){sample(c(0,1,2),1,prob=c(0.2,0.4,0.4))}

t_incub_fct_B <- function(x){rnorm(x,mean = 5,sd=1)}
p_max_fct_B <- function(x){rbeta(x,shape1 = 5,shape2=2)}

p_Exit_fct_B  <- function(t,current.in){
 if(current.in=="A"){return(0.1)}
 if(current.in=="B"){return(0.2)}
 if(current.in=="C"){return(1)}}

pTrans_hostB <- function(t,p_max,t_incub){
 if(t <= t_incub){p=0}
 if(t >= t_incub){p=p_max}
 return(p)
}

time_contact_B = function(t){round(rnorm(1, 3, 1), 0)}

transition.matrix = matrix(c(0,0.2,0.4,0.5,0,0.6,0.5,0.8,0),
                          nrow = 3, ncol = 3,
                          dimnames=list(c("A","B","C"),c("A","B","C")))

set.seed(6262)
test.nosoi <- nosoiSim(type="dual", popStructure="discrete",
                      length.sim=40,
                      max.infected.A=100,
                      max.infected.B=200,
                      init.individuals.A=1,
                      init.individuals.B=0,
                      init.structure.A="A",
                      init.structure.B=NA,
                      structure.matrix.A=transition.matrix,
                      structure.matrix.B=transition.matrix,
                      pExit.A = p_Exit_fctA,
                      param.pExit.A = list(t_infectA = t_infectA_fct),
                      pMove.A=p_Move_fctA,
                      param.pMove.A=NA,
                      timeDep.pMove.A=FALSE,
                      diff.pMove.A=FALSE,
                      timeDep.pExit.A=FALSE,
                      nContact.A = time_contact_A,
                      param.nContact.A = NA,
                      timeDep.nContact.A=FALSE,
                      pTrans.A = pTrans_hostA,
                      param.pTrans.A = list(t_infectA=t_infectA_fct),
                      timeDep.pTrans.A=FALSE,
                      prefix.host.A="H",
                      pExit.B = p_Exit_fct_B,
                      param.pExit.B = NA,
                      timeDep.pExit.B=FALSE,
                      diff.pExit.B=TRUE,
                      pMove.B=NA,
                      param.pMove.B=NA,
                      timeDep.pMove.B=FALSE,
                      diff.pMove.B=FALSE,
                      nContact.B = time_contact_B,
                      param.nContact.B = NA,
                      timeDep.nContact.B=FALSE,
                      pTrans.B = pTrans_hostB,
                      param.pTrans.B = list(p_max=p_max_fct_B,
                                            t_incub=t_incub_fct_B),
                      timeDep.pTrans.B=FALSE,
                      prefix.host.B="V")

test.nosoi