Package 'nosoi'

Title: A Forward Agent-Based Transmission Chain Simulator
Description: The aim of 'nosoi' (pronounced no.si) is to provide a flexible agent-based stochastic transmission chain/epidemic simulator (Lequime et al. Methods in Ecology and Evolution 11:1002-1007). It is named after the daimones of plague, sickness and disease that escaped Pandora's jar in the Greek mythology. 'nosoi' is able to take into account the influence of multiple variable on the transmission process (e.g. dual-host systems (such as arboviruses), within-host viral dynamics, transportation, population structure), alone or taken together, to create complex but relatively intuitive epidemiological simulations.
Authors: Sebastian Lequime [aut, cre] , Paul Bastide [aut] , Simon Dellicour [aut] , Philippe Lemey [aut] , Guy Baele [aut] , Thijs Janzen [ctb]
Maintainer: Sebastian Lequime <[email protected]>
License: GPL-3
Version: 1.1.2
Built: 2024-11-06 05:30:36 UTC
Source: https://github.com/slequime/nosoi

Help Index


Dual-host pathogen in structured (continuous) hosts populations

Description

This function runs a dual-host transmission chain simulation, with structured hosts populations (such as spatial features) in a shared continuous space. 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 passed. The movement of hosts on the continuous space map is a random walk (Brownian motion) that can be modified towards a biased random walk where hosts tend to be attracted to higher values of the environmental variable defined by the raster.

Usage

dualContinuous(
  length.sim,
  max.infected.A,
  max.infected.B,
  init.individuals.A,
  init.individuals.B,
  init.structure.A,
  init.structure.B,
  structure.raster.A,
  structure.raster.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,
  sdMove.A,
  param.sdMove.A,
  diff.sdMove.A = FALSE,
  timeDep.sdMove.A = FALSE,
  hostCount.sdMove.A = FALSE,
  attracted.by.raster.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,
  sdMove.B,
  param.sdMove.B,
  diff.sdMove.B = FALSE,
  timeDep.sdMove.B = FALSE,
  hostCount.sdMove.B = FALSE,
  attracted.by.raster.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
)

Arguments

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 location the initially infected host-A individuals are located. A vector of coordinates in the same coordinate space as the raster (NA if init.individual.A is 0).

init.structure.B

in which location the initially infected host-B individuals are located. A vector of coordinates in the same coordinate space as the raster (NA if init.individual.B is 0).

structure.raster.A

raster object defining the environmental variable for host-type A.

structure.raster.B

raster object defining the environmental variable for host 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

does pExit of host-type A depend on the environmental variable (set by the raster) (TRUE/FALSE).

hostCount.pExit.A

does pExit of host-type A vary with the host count (of either host-type A or B) in each raster cell? (TRUE/FALSE); if TRUE, 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

does pMove of host-type A depend on the environmental variable (set by the raster) (TRUE/FALSE).A.

hostCount.pMove.A

does pMove of host-type A vary with the host count (of either host-type A or B) in each raster cell? (TRUE/FALSE); if TRUE, diff.pMove.A should be TRUE.

sdMove.A

function that gives the distance traveled for host-type A (based on coordinates); output is the standard deviation value for the Brownian motion.

param.sdMove.A

parameter names (list of functions) for sdMove for host-type A.

diff.sdMove.A

does sdMove of host-type A depend on the environmental variable (set by the raster) (TRUE/FALSE).

timeDep.sdMove.A

is sdMove of host-type A dependent on the absolute time of the simulation (TRUE/FALSE) ?

hostCount.sdMove.A

does sdMove varies with the host count (of either host-type A or B) in each raster cell? (TRUE/FALSE); diff.sdMove.A should be TRUE.

attracted.by.raster.A

should the host-type A be attracted by higher values in the environmental raster? (TRUE/FALSE).

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

does nContact of host-type A depend on the environmental variable (set by the raster) (TRUE/FALSE).

hostCount.nContact.A

does nContact vary with the host count (of either host-type A or B) in each raster cell?? (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

does pTrans of host-type A depend on the environmental variable (set by the raster) (TRUE/FALSE).

hostCount.pTrans.A

does pTrans vary with the host count (of either host-type A or B) in each raster cell? (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

does pExit of host-type B depend on the environmental variable (set by the raster) (TRUE/FALSE).

hostCount.pExit.B

does pExit of host-type B vary with the host count (of either host-type A or B) in each raster cell? (TRUE/FALSE); if TRUE, 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

does pMove of host-type B depend on the environmental variable (set by the raster) (TRUE/FALSE).

hostCount.pMove.B

does pMove of host-type B vary with the host count (of either host-type A or B) in each raster cell? (TRUE/FALSE); if TRUE, diff.pMove.B should be TRUE.

sdMove.B

function that gives the distance traveled for host-type B (based on coordinates); output is the standard deviation value for the Brownian motion.

param.sdMove.B

parameter names (list of functions) for sdMove for host-type B.

diff.sdMove.B

does sdMove of host-type B depend on the environmental variable (set by the raster) (TRUE/FALSE).

timeDep.sdMove.B

is sdMove of host-type B dependent on the absolute time of the simulation (TRUE/FALSE) ?

hostCount.sdMove.B

does sdMove of host-type B vary with the host count (of either host-type A or B) in each raster cell? (TRUE/FALSE); if TRUE, diff.sdMove.B should be TRUE.

attracted.by.raster.B

should the host-type B be attracted by higher values in the environmental raster? (TRUE/FALSE)

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

does nContact of host-type B depend on the environmental variable (set by the raster) (TRUE/FALSE).

hostCount.nContact.B

does nContact of host-type B vary with the host count (of either host-type A or B) in each raster cell? (TRUE/FALSE); if TRUE, 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

does pTrans of host-type B depend on the environmental variable (set by the raster) (TRUE/FALSE).

hostCount.pTrans.B

does pTrans of host-type B vary with the host count (of either host-type A or B) in each raster cell? (TRUE/FALSE); if TRUE, 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.

Details

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").

Value

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

Raster

The structure raster(s) provided provided should of class raster. High values of the environmental variable can attract hosts if attracted.by.raster is TRUE. Raster have to share the same space (i.e. also the same cell size and ID).

Order of Arguments

The user specified function's arguments should follow this order: t (mandatory), prestime (optional, only if timeDep is TRUE), current.env.value (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.

Structure Parameters

The pMove function should return a single probability (a number between 0 and 1), and sdMove a real number (keep in mind this number is related to your coordinate space).

The use of diff (switch to TRUE) makes the corresponding function use the argument current.env.value (for "current environmental value").

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

Suffixes

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

See Also

For simulations with a discrete structure, see dualDiscrete. For simulations without any structures, see dualNone.

Examples

library(raster)

#Generating a raster for the movement
set.seed(860)

test.raster <- raster(nrows=100, ncols=100, xmn=-50, xmx=50, ymn=-50,ymx=50)
test.raster[] <- runif(10000, -80, 180)
test.raster <- focal(focal(test.raster, w=matrix(1, 5, 5), mean), w=matrix(1, 5, 5), mean)

t_incub_fct <- function(x){rnorm(x,mean = 5,sd=1)}
p_max_fct <- function(x){rbeta(x,shape1 = 5,shape2=2)}
p_Move_fct  <- function(t){return(0.1)}

sdMove_fct = function(t,current.env.value){return(100/(current.env.value+1))}

p_Exit_fct  <- function(t){return(0.08)}

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

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

start.pos <- c(0,0)

set.seed(805)
test.nosoi <- nosoiSim(type="dual", popStructure="continuous",
                      length.sim=200,
                      max.infected.A=500,
                      max.infected.B=500,
                      init.individuals.A=1,
                      init.individuals.B=0,
                      init.structure.A=start.pos,
                      init.structure.B=NA,
                      structure.raster.A=test.raster,
                      structure.raster.B=test.raster,
                      pExit.A=p_Exit_fct,
                      param.pExit.A=NA,
                      timeDep.pExit.A=FALSE,
                      diff.pExit.A=FALSE,
                      pMove.A=p_Move_fct,
                      param.pMove.A=NA,
                      timeDep.pMove.A=FALSE,
                      diff.pMove.A=FALSE,
                      diff.sdMove.A=TRUE,
                      sdMove.A=sdMove_fct,
                      param.sdMove.A=NA,
                      attracted.by.raster.A=TRUE,
                      nContact.A=time_contact,
                      param.nContact.A=NA,
                      timeDep.nContact.A=FALSE,
                      diff.nContact.A=FALSE,
                      pTrans.A=proba,
                      param.pTrans.A=list(p_max=p_max_fct,
                                          t_incub=t_incub_fct),
                      timeDep.pTrans.A=FALSE,
                      diff.pTrans.A=FALSE,
                      prefix.host.A="H",
                      pExit.B=p_Exit_fct,
                      param.pExit.B=NA,
                      timeDep.pExit.B=FALSE,
                      diff.pExit.B=FALSE,
                      pMove.B=p_Move_fct,
                      param.pMove.B=NA,
                      timeDep.pMove.B=FALSE,
                      diff.pMove.B=FALSE,
                      diff.sdMove.B=TRUE,
                      sdMove.B=sdMove_fct,
                      param.sdMove.B=NA,
                      attracted.by.raster.B=TRUE,
                      nContact.B=time_contact,
                      param.nContact.B=NA,
                      timeDep.nContact.B=FALSE,
                      diff.nContact.B=FALSE,
                      pTrans.B=proba,
                      param.pTrans.B=list(p_max=p_max_fct,
                                          t_incub=t_incub_fct),
                      timeDep.pTrans.B=FALSE,
                      diff.pTrans.B=FALSE,
                      prefix.host.B="V")
test.nosoi

Dual-host pathogen in structured (discrete) hosts populations

Description

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.

Usage

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
)

Arguments

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.

Details

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").

Value

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

Structure Matrix

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.

Order of Arguments

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.

Structure Parameters

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.

Suffixes

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

See Also

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

Examples

#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

Dual-host pathogen in homogeneous hosts populations

Description

This function, that can be wrapped within nosoiSim, runs a dual-host transmission chain simulation, without any structure features in both hosts populations. 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.

Usage

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

Arguments

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).

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)?

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)?

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)?

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)?

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)?

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)?

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.

Details

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").

Value

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

Suffixes

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

Order of Arguments

The user specified function's arguments should follow this order: t (mandatory), prestime (optional, only if timeDep is TRUE), parameters specified in the list.

See Also

For simulations with a discrete structured host population, see dualDiscrete. For simulations with a structured population in continuous space, dualContinuous

Examples

#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_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))}

#Host B
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,prestime){(sin(prestime/12)+1)/5}

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)}

set.seed(90)
test.nosoi <- nosoiSim(type="dual", popStructure="none",
                       length.sim=40,
                       max.infected.A=100,
                       max.infected.B=200,
                       init.individuals.A=1,
                       init.individuals.B=0,
                       pExit.A = p_Exit_fctA,
                       param.pExit.A = list(t_infectA = t_infectA_fct),
                       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=TRUE,
                       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

Gets the cumulative number of infected hosts for the full length of the simulation

Description

This function computes from the output of a nosoiSim simulation the cumulative count of infected hosts at each time step of the simulation. The output is a data.table.

Usage

getCumulative(nosoi.output)

Arguments

nosoi.output

Output of a nosoi simulation (object of class nosoiSim).

Value

The output is a data.table with the following structure:

t

Time-step (integer).

Count

Cumulative number of infected hosts at given time-step.

type

Host-type, identified by its user-defined prefix.

See Also

summary.nosoiSim


Gets the current number of infected hosts for the full length of the simulation

Description

This function computes from the output of a nosoiSim simulation the dynamic count of infected hosts at each time step (and each state if discrete structure) of the simulation. The output is a data.table.

Usage

getDynamic(nosoi.output)

Arguments

nosoi.output

Output of a nosoi simulation (object of class nosoiSim).

Value

The output is a data.table with the following structure:

state

(only when discrete structure) Given state

Count

Current number of infected hosts at given time-step.

type

Host-type, identified by its user-defined prefix.

t

Time-step (integer).

See Also

summary.nosoiSim


Extracts specific data from a nosoiSim object

Description

This function extracts data user-defined data (i.e. table.hosts, N.infected, table.state or popStructure) from a nosoiSim object.

Usage

getHostData(
  nosoi.output,
  what = c("table.hosts", "N.infected", "table.state", "popStructure"),
  pop = "A"
)

Arguments

nosoi.output

an object of class nosoiSim

what

the data to get, among table.hosts, N.infected, table.state or popStructure.

pop

the population to be extracted (one of "A" or "B")

Value

Returns a data.table with the requested data.

See Also

To directly extract table.hosts or table.state, you can also use getTableHosts and getTableState respectively.

Examples

t_incub_fct <- function(x){rnorm(x,mean = 5,sd=1)}
p_max_fct <- function(x){rbeta(x,shape1 = 5,shape2=2)}
p_Exit_fct  <- function(t){return(0.08)}

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

time_contact <- function(t){round(rnorm(1, 3, 1), 0)}

test.nosoi <- nosoiSim(type="single", popStructure="none",
                       length=40,
                       max.infected=100,
                       init.individuals=1,
                       nContact=time_contact,
                       param.nContact=NA,
                       pTrans = proba,
                       param.pTrans = list(p_max=p_max_fct,
                                           t_incub=t_incub_fct),
                       pExit=p_Exit_fct,
                       param.pExit=NA)


data.extracted <- getHostData(test.nosoi, "table.hosts", "A")

Gets R0 from a nosoi simulation

Description

Gets an estimate of secondary cases (what R0 usually tries to estimate) and its distribution from the output of a nosoiSim simulation. The actual calculation is based on inactive hosts at the end of the simulation to avoid bias introduced by hosts that have not finished their transmission potential.

Usage

getR0(nosoi.output)

Arguments

nosoi.output

Output of a nosoi simulation (object of class nosoiSim).

Details

Current getR0 (after and including version 1.1.0) is a corrected version. In previous versions (prior to 1.1.0), the output included in its computation hosts that should not have been counted (still active).

Value

A list with the following items:

N.inactive

Number of inactive hosts at the end of the simulation.

R0.mean

Mean R0 based on the distribution (see below).

R0.dist

Distribution for each host of the secondary cases it generated (in case of dual-hosts, then the secondary cases of the same host-type).

See Also

summary.nosoiSim


Extracts table.hosts from a nosoiSim object

Description

This function extracts the table.hosts for the request host-type from a nosoiSim object.

Usage

getTableHosts(nosoi.output, pop = "A")

Arguments

nosoi.output

an object of class nosoiSim

pop

the host-type to be extracted (either "A" or "B", if not dual-host, then "A")

Value

Returns a data.table with the requested data. The table.hosts (class data.table) contains informations about each host that has been simulated (one row is one host). The structure of the table is the following:

hosts.ID

Unique identifier for the host, based on user-defined prefix and an integer.

inf.by

Unique identifier for the host that infected the current one.

inf.in

(only if structure is present) State or coordinates (in that case inf.in.x and inf.in.y) in which the host was infected.

current.in

(only if structure is present) State or coordinates (in that case current.in.x and current.in.y) in which the host is at the end of the simulation.

current.env.value

(only if continuous structure is present) Environmental value (raster cell value) in which the host is at the end of the simulation.

current.cell.raster

(only if continuous structure is present) Raster cell numeric ID in which the host is at the end of the simulation.

host.count

(only if structure is present) Host count in the current state or raster cell (beware, updated only if used).

inf.time

When did the host enter the simulation (infection time).

out.time

When did the host exit the simulation (NA if still active).

active

Is the host still active at the end of the simulation (TRUE for YES, FALSE for NO)?

parameters

The remaining columns are the sampled values for the individual-based parameters (if any) specified by the user.


Extracts table.state from a nosoiSim object

Description

This function extracts the table.state for the request host-type from a nosoiSim object. table.state is present only if there is any structure (discrete or continuous) used.

Usage

getTableState(nosoi.output, pop = "A")

Arguments

nosoi.output

an object of class nosoiSim

pop

the host-type to be extracted (either "A" or "B", if not dual-host, then "A")

Value

Returns a data.table with the requested data. The table.state (class data.table) contains informations the location of each host during time (one row is one host at one location). The structure of the table is the following:

hosts.ID

Unique identifier for the host, based on user-defined prefix and an integer.

state

State or coordinates (in that case state.x and state.y) in which the host is during that time interval.

current.env.value

(only if continuous structure is present) Environmental value (raster cell value) in which the host is at the end of the simulation.

current.cell.raster

(only if continuous structure is present) Raster cell numeric ID in which the host is at the end of the simulation.

time.from

Time-step at which the host moved to the location.

time.to

Time-step at which the host exited the location (either by exiting the simulation or moving somewhere else).


Gets the full transmission tree (phylogenetic tree-like) from a nosoi simulation

Description

From a nosoi simulated epidemics, this function extracts the full transmission tree in a form mimicking a phylogenetic tree.

Usage

getTransmissionTree(nosoiInf)

Arguments

nosoiInf

an object of class nosoiSim

Details

This function uses packages tidytree and treeio, that rely on ape.

Value

A tree of class treedata, containing a phylogenetic tree based on the transmission chain and the mapped data at all the nodes.

See Also

For exporting the annotated tree to other software packages, see functions in treeio (e.g. write.beast).

To sub-sample this tree, see functions sampleTransmissionTree and sampleTransmissionTreeFromExiting

Examples

t_incub_fct <- function(x){rnorm(x,mean = 5,sd=1)}
p_max_fct <- function(x){rbeta(x,shape1 = 5,shape2=2)}
p_Exit_fct  <- function(t){return(0.08)}
p_Move_fct  <- function(t){return(0.1)}

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

time_contact = 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(805)
test.nosoi <- nosoiSim(type="single", popStructure="discrete",
                        length=20,
                        max.infected=100,
                        init.individuals=1,
                        init.structure="A",
                        structure.matrix=transition.matrix,
                        pMove=p_Move_fct,
                        param.pMove=NA,
                        nContact=time_contact,
                        param.nContact=NA,
                        pTrans = proba,
                        param.pTrans = list(p_max=p_max_fct,
                                            t_incub=t_incub_fct),
                        pExit=p_Exit_fct,
                        param.pExit=NA
)

## Make sure all needed packages are here
if (requireNamespace("ape", quietly = TRUE) &&
    requireNamespace("tidytree", quietly = TRUE) &&
    requireNamespace("treeio", quietly = TRUE)) {
  library(ape)
  library(tidytree)
  library(treeio)

  #' ## Full transmission tree
  ttreedata <- getTransmissionTree(test.nosoi)
  plot(ttreedata@phylo)

  ## Sampling "non dead" individuals
  hID <- c("H-1", "H-7", "H-15", "H-100")
  samples <- data.table(hosts = hID,
                        times = c(5.2, 9.3, 10.2, 16),
                        labels = paste0(hID, "-s"))

  sampledTree <- sampleTransmissionTree(test.nosoi, ttreedata, samples)
  plot(sampledTree@phylo)

  ## Sampling "dead" individuals
  sampledDeadTree <- sampleTransmissionTreeFromExiting(ttreedata, hID)
  plot(sampledDeadTree@phylo)
  }

Top-level function to use nosoi.

Description

This function determines which general settings the user wants to use for his simulation. All other arguments are passed down to the chosen simulator itself, such as singleNone, singleDiscrete, singleContinuous, dualNone, dualDiscrete or dualContinuous.

Usage

nosoiSim(type = "single", popStructure = "none", ...)

Arguments

type

specifies which type of pathogen we are interested in, either "single" or "dual"-host (e.g. arboviruses).

popStructure

specifies if the population in which the transmission is to occur is structured ("none", "discrete" or "continuous").

...

arguments to be passed on to the chosen simulator itself, such as singleNone, singleDiscrete, singleContinuous, dualNone, dualDiscrete or dualContinuous.

Value

An object of class nosoiSim, containing all results of the simulation. Class nosoiSim object have the following slots:

total.time

Number of time steps the simulation ran (integer).

type

String giving the simulation type ("single" or "dual" host).

host.info.A: object of class nosoiSimOne
N.infected

Number of infected hosts (integer).

table.hosts

Table containing the results of the simulation (see getTableHosts for more details on the table).

table.state

Table containing the results of the simulation, focusing on the movement history of each host (see getTableState for more details on the table).

prefix.host

String containing the prefix used to name hosts (character string).

popStructure

String giving the population structure (one of "none", "discrete" or "continuous").

host.info.B: object of class nosoiSimOne

Same structure as host.info.A, but for host B (if it exists).

See Also

Individual simulation functions:

singleNone, singleDiscrete, singleContinuous, dualNone, dualDiscrete and dualContinuous.

Functions to extract the results:

getTableHosts, getTableState

Summary statistics functions:

nosoiSummary, getCumulative, getDynamic, getR0

Examples

t_incub_fct <- function(x){rnorm(x,mean = 5,sd=1)}
p_max_fct <- function(x){rbeta(x,shape1 = 5,shape2=2)}
p_Exit_fct  <- function(t){return(0.08)}

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

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

test.nosoi <- nosoiSim(type="single", popStructure="none",
                      length=40,
                      max.infected=100,
                      init.individuals=1,
                      nContact=time_contact,
                      param.nContact=NA,
                      pTrans = proba,
                      param.pTrans = list(p_max=p_max_fct,
                                          t_incub=t_incub_fct),
                      pExit=p_Exit_fct,
                      param.pExit=NA)
test.nosoi

Summarizes the epidemiological features of a nosoi simulation

Description

This function provides summary information about the simulation (number of infected hosts, R0, etc.) as a list.

Usage

nosoiSummary(object)

## S3 method for class 'nosoiSim'
summary(object, ...)

Arguments

object

Output of a nosoi simulation (object of class nosoiSim).

...

further arguments passed to or from other methods.

Value

All computed data is provided in a list:

R0

Provides a sublist with number of inactive hosts at the end of the simulation N.inactive, mean R0 R0.mean, and R0 distribution R0.dist. For more details, see getR0.

dynamics

data.table with the count of currently infected (i.e. active) hosts at each time step of the simulation (by state if the simulation was in a discrete structured host population). For more details, see getDynamic.

cumulative

data.table with the cumulative count of infected hosts at each time step of the simulation. For more details, see getCumulative.

See Also

You can directly compute each elements of the list without using the summarise function. See getR0, getDynamic and getCumulative.

Examples

t_incub_fct <- function(x){rnorm(x,mean = 5,sd=1)}
p_max_fct <- function(x){rbeta(x,shape1 = 5,shape2=2)}
p_Exit_fct  <- function(t){return(0.08)}

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

time_contact <- function(t){round(rnorm(1, 3, 1), 0)}

test.nosoi <- nosoiSim(type="single", popStructure="none",
                      length=40,
                      max.infected=100,
                      init.individuals=1,
                      nContact=time_contact,
                      param.nContact=NA,
                      pTrans = proba,
                      param.pTrans = list(p_max=p_max_fct,
                                          t_incub=t_incub_fct),
                      pExit=p_Exit_fct,
                      param.pExit=NA)


nosoiSummary(test.nosoi)

Sample the transmission tree (phylogenetic tree-like)

Description

Sample a full transmission tree. This function allows for sampling multiple times on the same lineage. When this happens, the sampled ancestor is a tip with length zero.

Usage

sampleTransmissionTree(nosoiInf, tree, samples)

Arguments

nosoiInf

an object of class nosoiSim

tree

a treedata object created by function getTransmissionTree

samples

a data.table object with the following entries:

hosts

Host ID of the individuals to be sampled

times

Times at which each host is sampled

labels

label for the corresponding tip in the tree

Details

The tree needs to be produced by function getTransmissionTree applied on the same nosoiSim object.

Value

A tree of class treedata, containing a phylogenetic tree based on the transmission chain and the mapped data at all the nodes.

See Also

For exporting the annotated tree to other software packages, see functions in treeio (e.g. write.beast).

To get the full transmission matrix, see getTransmissionTree.

For sampling only dead individuals, see sampleTransmissionTreeFromExiting.

Examples

t_incub_fct <- function(x){rnorm(x,mean = 5,sd=1)}
p_max_fct <- function(x){rbeta(x,shape1 = 5,shape2=2)}
p_Exit_fct  <- function(t){return(0.08)}
p_Move_fct  <- function(t){return(0.1)}

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

time_contact = 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(805)
test.nosoi <- nosoiSim(type="single", popStructure="discrete",
                        length=20,
                        max.infected=100,
                        init.individuals=1,
                        init.structure="A",
                        structure.matrix=transition.matrix,
                        pMove=p_Move_fct,
                        param.pMove=NA,
                        nContact=time_contact,
                        param.nContact=NA,
                        pTrans = proba,
                        param.pTrans = list(p_max=p_max_fct,
                                            t_incub=t_incub_fct),
                        pExit=p_Exit_fct,
                        param.pExit=NA
)

## Make sure all needed packages are here
if (requireNamespace("ape", quietly = TRUE) &&
    requireNamespace("tidytree", quietly = TRUE) &&
    requireNamespace("treeio", quietly = TRUE)) {
  library(ape)
  library(tidytree)
  library(treeio)

  #' ## Full transmission tree
  ttreedata <- getTransmissionTree(test.nosoi)
  plot(ttreedata@phylo)

  ## Sampling "non dead" individuals
  hID <- c("H-1", "H-7", "H-15", "H-100")
  samples <- data.table(hosts = hID,
                        times = c(5.2, 9.3, 10.2, 16),
                        labels = paste0(hID, "-s"))

  sampledTree <- sampleTransmissionTree(test.nosoi, ttreedata, samples)
  plot(sampledTree@phylo)

  ## Sampling "dead" individuals
  sampledDeadTree <- sampleTransmissionTreeFromExiting(ttreedata, hID)
  plot(sampledDeadTree@phylo)
  }

Sample the transmission tree (phylogenetic tree-like) among the exited hosts

Description

Sample a full transmission tree. This function allows for sampling only exited (i.e. inactive) individuals (e.g. when the sampling procedure is destructive or cuts the hosts from the population). Beware because it does not influence the epidemiological process, it only means that the host has been sampled when exiting the simulation.

Usage

sampleTransmissionTreeFromExiting(tree, hosts)

Arguments

tree

a treedata object created by function getTransmissionTree

hosts

a vector of dead hosts to sample

Details

The tree needs to be produced by function getTransmissionTree applied on the same nosoiSim object.

Value

A tree of class treedata, containing a phylogenetic tree based on the transmission chain and the mapped data at all the nodes.

See Also

For exporting the annotated tree to other software packages, see functions in treeio (e.g. write.beast).

To get the full transmission matrix, see getTransmissionTree.

For sampling among non-dead individuals, see sampleTransmissionTree.

Examples

t_incub_fct <- function(x){rnorm(x,mean = 5,sd=1)}
p_max_fct <- function(x){rbeta(x,shape1 = 5,shape2=2)}
p_Exit_fct  <- function(t){return(0.08)}
p_Move_fct  <- function(t){return(0.1)}

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

time_contact = 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(805)
test.nosoi <- nosoiSim(type="single", popStructure="discrete",
                        length=20,
                        max.infected=100,
                        init.individuals=1,
                        init.structure="A",
                        structure.matrix=transition.matrix,
                        pMove=p_Move_fct,
                        param.pMove=NA,
                        nContact=time_contact,
                        param.nContact=NA,
                        pTrans = proba,
                        param.pTrans = list(p_max=p_max_fct,
                                            t_incub=t_incub_fct),
                        pExit=p_Exit_fct,
                        param.pExit=NA
)

## Make sure all needed packages are here
if (requireNamespace("ape", quietly = TRUE) &&
    requireNamespace("tidytree", quietly = TRUE) &&
    requireNamespace("treeio", quietly = TRUE)) {
  library(ape)
  library(tidytree)
  library(treeio)

  #' ## Full transmission tree
  ttreedata <- getTransmissionTree(test.nosoi)
  plot(ttreedata@phylo)

  ## Sampling "non dead" individuals
  hID <- c("H-1", "H-7", "H-15", "H-100")
  samples <- data.table(hosts = hID,
                        times = c(5.2, 9.3, 10.2, 16),
                        labels = paste0(hID, "-s"))

  sampledTree <- sampleTransmissionTree(test.nosoi, ttreedata, samples)
  plot(sampledTree@phylo)

  ## Sampling "dead" individuals
  sampledDeadTree <- sampleTransmissionTreeFromExiting(ttreedata, hID)
  plot(sampledDeadTree@phylo)
  }

Single-host pathogen in a structured (continuous) host population

Description

This function runs a single-host transmission chain simulation, with a structured host population (such as spatial features) in a continuous space. 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 passed. The movement of hosts on the continuous space map is a random walk (Brownian motion) that can be modified towards a biased random walk where hosts tend to be attracted to higher values of the environmental variable defined by the raster.

Usage

singleContinuous(
  length.sim,
  max.infected,
  init.individuals,
  init.structure,
  structure.raster,
  diff.pExit = FALSE,
  timeDep.pExit = FALSE,
  hostCount.pExit = FALSE,
  pExit,
  param.pExit,
  diff.pMove = FALSE,
  timeDep.pMove = FALSE,
  hostCount.pMove = FALSE,
  pMove,
  param.pMove,
  diff.sdMove = FALSE,
  timeDep.sdMove = FALSE,
  hostCount.sdMove = FALSE,
  sdMove,
  param.sdMove,
  attracted.by.raster = FALSE,
  diff.nContact = FALSE,
  timeDep.nContact = FALSE,
  hostCount.nContact = FALSE,
  nContact,
  param.nContact,
  diff.pTrans = FALSE,
  timeDep.pTrans = FALSE,
  hostCount.pTrans = FALSE,
  pTrans,
  param.pTrans,
  prefix.host = "H",
  print.progress = TRUE,
  print.step = 10
)

Arguments

length.sim

specifies the length (in unit of time) over which the simulation should be run.

max.infected

specifies the maximum number of hosts that can be infected in the simulation.

init.individuals

number of initially infected individuals.

init.structure

in which location the initially infected individuals are located. A vector of coordinates in the same coordinate space as the raster.

structure.raster

raster object defining the environmental variable.

diff.pExit

does pExit depend on the environmental variable (set by the raster) (TRUE/FALSE).

timeDep.pExit

is pExit dependent on the absolute time of the simulation? (TRUE/FALSE)

hostCount.pExit

does pExit vary with the host count in each raster cell? (TRUE/FALSE); if TRUE, diff.pExit should be TRUE.

pExit

function that gives the probability to exit the simulation for an infected host (either moving out, dying, etc.).

param.pExit

parameter names (list of functions) for the pExit.

diff.pMove

does pMove depend on the environmental variable (set by the raster) (TRUE/FALSE).

timeDep.pMove

does pMove depend on the absolute time of the simulation (TRUE/FALSE).

hostCount.pMove

does pMove vary with the host count in each raster cell? (TRUE/FALSE); if TRUE, diff.pMove should also be TRUE.

pMove

function that gives the probability of a host moving as a function of time.

param.pMove

parameter names (list of functions) for the pMove.

diff.sdMove

does sdMove depend on the environmental variable (set by the raster) (TRUE/FALSE).

timeDep.sdMove

does sdMove depend on the absolute time of the simulation (TRUE/FALSE).

hostCount.sdMove

does sdMove vary with the host count in each raster cell? (TRUE/FALSE); if TRUE, diff.sdMove should be TRUE.

sdMove

function that gives the distance traveled (based on coordinates); output is the standard deviation value for the Brownian motion.

param.sdMove

parameter names (list of functions) for sdMove.

attracted.by.raster

should the hosts be attracted by higher values in the environmental raster? (TRUE/FALSE).

diff.nContact

does nContact depend on the environmental variable (set by the raster) (TRUE/FALSE).

timeDep.nContact

is nContact dependent on the absolute time of the simulation? (TRUE/FALSE)

hostCount.nContact

does nContact vary with the host count in each raster cell? (TRUE/FALSE); if TRUE, diff.nContact should be TRUE.

nContact

function that gives the number of potential transmission events per unit of time.

param.nContact

parameter names (list of functions) for param.nContact.

diff.pTrans

does pTrans depend on the environmental variable (set by the raster) (TRUE/FALSE).

timeDep.pTrans

is pTrans dependent on the absolute time of the simulation? (TRUE/FALSE)

hostCount.pTrans

does pTrans vary with the host count in each raster cell? (TRUE/FALSE); if TRUE, diff.pTrans should be TRUE.

pTrans

function that gives the probability of transmit a pathogen as a function of time since infection.

param.pTrans

parameter names (list of functions) for the pExit.

prefix.host

character(s) to be used as a prefix for the hosts 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.

Details

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").

Value

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

Raster

The structure raster provided provided should of class raster. High values of the environmental variable can attract hosts if attracted.by.raster is TRUE.

Structure Parameters

The pMove function should return a single probability (a number between 0 and 1), and sdMove a real number (keep in mind this number is related to your coordinate space).

The use of diff (switch to TRUE) makes the corresponding function use the argument current.env.value (for "current environmental value").

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

Order of Arguments

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

See Also

For simulations with a discrete structure, see singleDiscrete. For simulations without any structures, see singleNone.

Examples

library(raster)
#Generating a raster for the movement
set.seed(860)

test.raster <- raster(nrows=100, ncols=100, xmn=-50, xmx=50, ymn=-50,ymx=50)
test.raster[] <- runif(10000, -80, 180)
test.raster <- focal(focal(test.raster, w=matrix(1, 5, 5), mean), w=matrix(1, 5, 5), mean)
plot(test.raster)

t_incub_fct <- function(x){rnorm(x,mean = 5,sd=1)}
p_max_fct <- function(x){rbeta(x,shape1 = 5,shape2=2)}
p_Move_fct  <- function(t){return(0.1)}

sdMove_fct = function(t,current.env.value){return(100/(current.env.value+1))}

p_Exit_fct  <- function(t){return(0.08)}

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

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

start.pos <- c(0,0)

test.nosoiA <- nosoiSim(type="single", popStructure="continuous",
                   length=200,
                   max.infected=500,
                   init.individuals=1,
                   init.structure=start.pos,
                   structure.raster=test.raster,
                   pMove=p_Move_fct,
                   param.pMove=NA,
                   diff.sdMove=TRUE,
                   sdMove=sdMove_fct,
                   param.sdMove=NA,
                   attracted.by.raster=TRUE,
                   nContact=time_contact,
                   param.nContact=NA,
                   pTrans = proba,
                   param.pTrans = list(p_max=p_max_fct,
                                       t_incub=t_incub_fct),
                   pExit=p_Exit_fct,
                   param.pExit=NA)

Single-host pathogen in a structured (discrete) host population

Description

This function, that can be wrapped within nosoiSim, runs a single-host transmission chain simulation, with a discrete host population structure (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.

Usage

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

Arguments

length.sim

specifies the length (in unit of time) over which the simulation should be run.

max.infected

specifies the maximum number of hosts that can be infected in the simulation.

init.individuals

number of initially infected individuals.

init.structure

in which state (e.g. location) the initially infected individuals are located.

structure.matrix

transition matrix (probabilities) to go from location A (row) to B (column)

diff.pExit

is pExit different between states of the structured population (TRUE/FALSE)

timeDep.pExit

is pExit dependent on the absolute time of the simulation? (TRUE/FALSE)

hostCount.pExit

does pExit varies with the host count in the state? (TRUE/FALSE); diff.pExit should be TRUE.

pExit

function that gives the probability to exit the simulation for an infected host (either moving out, dying, etc.).

param.pExit

parameter names (list of functions) for the pExit.

diff.pMove

is pMove different between states of the structured population (TRUE/FALSE)

timeDep.pMove

is pMove dependent on the absolute time of the simulation (TRUE/FALSE)

hostCount.pMove

does pMove varies with the host count in the state? (TRUE/FALSE); diff.pMove should be TRUE.

pMove

function that gives the probability of a host moving as a function of time.

param.pMove

parameter names (list of functions) for the pMove.

diff.nContact

is nContact different between states of the structured population (TRUE/FALSE)

timeDep.nContact

is nContact dependent on the absolute time of the simulation? (TRUE/FALSE)

hostCount.nContact

does nContact varies with the host count in the state? (TRUE/FALSE); diff.nContact should be TRUE.

nContact

function that gives the number of potential transmission events per unit of time.

param.nContact

parameter names (list of functions) for param.nContact.

diff.pTrans

is pTrans different between states of the structured population (TRUE/FALSE)

timeDep.pTrans

is pTrans dependent on the absolute time of the simulation? (TRUE/FALSE)

hostCount.pTrans

does pTrans varies with the host count in the state? (TRUE/FALSE); diff.pTrans should be TRUE.

pTrans

function that gives the probability of transmit a pathogen as a function of time since infection.

param.pTrans

parameter names (list of functions) for the pExit.

prefix.host

character(s) to be used as a prefix for the hosts 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.

Details

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").

Value

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

Structure Matrix

The structure 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.

Structure Parameters

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.

Order of Arguments

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 (optional, only if hostCount is TRUE) and parameters specified in the list.

See Also

For simulations with a structure in continuous space, see singleContinuous. For simulations without any structures, see singleNone.

Examples

t_incub_fct <- function(x){rnorm(x,mean = 5,sd=1)}
p_max_fct <- function(x){rbeta(x,shape1 = 5,shape2=2)}
p_Exit_fct  <- function(t){return(0.08)}
p_Move_fct  <- function(t){return(0.1)}

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

time_contact = 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(805)
test.nosoiA <- nosoiSim(type="single", popStructure="discrete",
                       length=20,
                       max.infected=100,
                       init.individuals=1,
                       init.structure="A",
                       structure.matrix=transition.matrix,
                       pMove=p_Move_fct,
                       param.pMove=NA,
                       nContact=time_contact,
                       param.nContact=NA,
                       pTrans = proba,
                       param.pTrans = list(p_max=p_max_fct,
                                           t_incub=t_incub_fct),
                      pExit=p_Exit_fct,
                      param.pExit=NA)

Single-host pathogen in a homogeneous host population

Description

This function, that can be wrapped within nosoiSim, runs a single-host transmission chain simulation, without any structure features in the host population. 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.

Usage

singleNone(
  length.sim,
  max.infected,
  init.individuals,
  pExit,
  param.pExit,
  timeDep.pExit = FALSE,
  nContact,
  param.nContact,
  timeDep.nContact = FALSE,
  pTrans,
  param.pTrans,
  timeDep.pTrans = FALSE,
  prefix.host = "H",
  print.progress = TRUE,
  print.step = 10
)

Arguments

length.sim

specifies the length (in unit of time) over which the simulation should be run.

max.infected

specifies the maximum number of hosts that can be infected in the simulation.

init.individuals

number of initially infected individuals.

pExit

function that gives the probability to exit the simulation for an infected host (either moving out, dying, etc.).

param.pExit

parameter names (list of functions) for the pExit.

timeDep.pExit

is pExit dependent on the absolute time of the simulation? (TRUE/FALSE)

nContact

function that gives the number of potential transmission events per unit of time.

param.nContact

parameter names (list of functions) for param.nContact.

timeDep.nContact

is nContact dependent on the absolute time of the simulation? (TRUE/FALSE)

pTrans

function that gives the probability of transmit a pathogen as a function of time since infection.

param.pTrans

parameter names (list of functions) for the pExit.

timeDep.pTrans

is pTrans dependent on the absolute time of the simulation? (TRUE/FALSE)

prefix.host

character(s) to be used as a prefix for the hosts 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.

Details

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").

Value

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

Order of Arguments

The user specified function's arguments should follow this order: t (mandatory), prestime (optional, only if timeDep is TRUE), parameters specified in the list.

See Also

For simulations with a discrete structured host population, see singleDiscrete. For simulations with a structured population in continuous space, singleContinuous

Examples

t_incub_fct <- function(x){rnorm(x,mean = 5,sd=1)}
p_max_fct <- function(x){rbeta(x,shape1 = 5,shape2=2)}
p_Exit_fct  <- function(t){return(0.08)}

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

time_contact <- function(t){round(rnorm(1, 3, 1), 0)}

test.nosoi <- nosoiSim(type="single", popStructure="none",
                      length=40,
                      max.infected=100,
                      init.individuals=1,
                      nContact=time_contact,
                      param.nContact=NA,
                      pTrans = proba,
                      param.pTrans = list(p_max=p_max_fct,
                                          t_incub=t_incub_fct),
                      pExit=p_Exit_fct,
                      param.pExit=NA)

test.nosoi