Package 'rescomp'

Title: Efficient Modelling of Resource Competition
Description: Generate, simulate and visualise ODE models of consumer-resource interactions. At its core, 'rescomp' provides a resource competition modelling focused interface to 'deSolve', alongside flexible functions for visualising model properties and dynamics. More information, documentation and examples can be found on the package website.
Authors: Christopher R P Brown [aut, cre] (ORCID: <https://orcid.org/0000-0002-0393-3477>), Jan Engelstaedter [aut] (ORCID: <https://orcid.org/0000-0003-3340-918X>), Andrew Letten [aut] (ORCID: <https://orcid.org/0000-0001-6436-7942>)
Maintainer: Christopher R P Brown <[email protected]>
License: MIT + file LICENSE
Version: 1.0.0
Built: 2026-06-02 07:32:14 UTC
Source: https://github.com/pyrrhicpachyderm/rescomp

Help Index

Applies a rescomp_event object to modify state variables

Description

Applies the instantaneous changes specified by a rescomp_event object to the full set of state variables (species and resources).

Usage

apply_event(event_obj, species, resources, params, time)

Arguments

event_obj

An object of class rescomp_event.
species

A vector of species concentrations.
resources

A vector of resource concentrations.
params

A list of time-dependent parameters.
time

The current simulation time.

Details

This function is normally only for internal use, but is exported to aid users in debugging their created rescomp_event objects.

Value

A vector of state variables, species concentrations followed by resource concentrations.

Examples

batch_transfer <- event_batch_transfer(dilution = 0.1, resources = c(1, 1))
apply_event(
  batch_transfer,
  species = c(100, 200),
  resources = c(0.5, 2),
  params = list(),
  time = 0 # Not used by batch transfer events.
)

Shorthand to create a matrix with spnum rows and resnum columns

Description

This produces an object that, when passed through rescomp functions such as funcresp_type1(), and ultimately provided to spec_rescomp(), inherits spnum and resnum from the arguments to spec_rescomp(). Thus, it is shorthand for matrix(c(...), nrow = spnum, ncol = resnum, byrow = TRUE). Inheritance only works when passed directly through rescomp functions; it will fail if used in the func provided to a function such as funcresp_custom().

Usage

crmatrix(..., byrow = TRUE)

Arguments

...

One or more numeric vectors that, concatenated by c(), produce a vector of length spnum * resnum.
byrow

A logical vector of length 1, equivalent to byrow of matrix(). Defaults to TRUE, unlike the byrow argument of matrix().

Value

S3 object of class rescomp_crmatrix.

Examples

pars <- spec_rescomp(
  spnum = 3, resnum = 2,
  funcresp = funcresp_type1(crmatrix(
    0.10, 0.10,
    0.08, 0.12,
    0.15, 0.05
  ))
)
plot_rescomp(sim_rescomp(pars))

Define consumer resource ODE function

Description

Define consumer resource ODE function

Usage

def_cr_ode(t, y, pars)

Arguments

t

The current time of the simulation.
y

The vector of current estimates of consumers and resources in the simulation.
pars

S3 object of class rescomp containing model specification.

Value

List of length one, containing a vector of derivatives of y with respect to time.

Define an event in which the species and resources are diluted by addition of new medium

Description

Produces an event object representing a batch transfer by dilution with new medium. First, all existing species and resource concentrations are multiplied by dilution. Then, the resource concentrations listed in resources, multiplied by 1 - dilution, are added.

Usage

event_batch_transfer(dilution, resources)

Arguments

dilution

A numeric vector or rescomp_coefs_vector of length 1 (whose value should be between 0 and 1) representing the proportion of the original medium retained.
resources

A numeric vector or rescomp_coefs_vector of resource concentrations.

Value

S3 object of class rescomp_event.

Examples

m1 <- spec_rescomp(
  events = list(
    event_schedule_periodic(
      event_batch_transfer(dilution = 0.1, resources = 1),
      period = 250
    )
  )
)
plot_rescomp(sim_rescomp(m1))

Define an event in which a constant amount of some resources is added to (or removed from) the system

Description

Produces an event object representing a pulse of added/removed resources.

Usage

event_res_add(resources, min_zero = TRUE)

Arguments

resources

A numeric vector or rescomp_coefs_vector of resource concentrations, by which the current resource concentrations are increased. Can be negative, to decrease resource concentrations.
min_zero

If this is TRUE, resulting resource concentrations are clamped to a minimum of zero. If this is FALSE, negative values of resources may reduce resource concentrations below zero.

Value

S3 object of class rescomp_event.

Examples

m1 <- spec_rescomp(
  spnum = 2, resnum = 2,
  funcresp = funcresp_type1(crmatrix(
    0.10, 0.08,
    0.06, 0.12
  )),
  events = list(
    event_schedule_periodic(
      event_res_add(c(-0.1, 0.1)),
      period = 250
    )
  )
)
plot_rescomp(sim_rescomp(m1))

Define an event which instantaneously changes resource concentrations using an arbitrary function

Description

Produces an event object suitable for building event schedules to pass spec_rescomp().

Usage

event_res_custom(func, resnum = NULL)

Arguments

func

A function that takes resources (a numeric vector of resource concentrations) and params (a list of parameters) and returns a numeric vector of resource concentrations.
resnum

The number of resources.

Details

If resnum is NULL, spec_rescomp() will attempt to infer it. This is fine if the result is used directly in an event schedule, but may fail if the result must be combined with other rescomp_events first.

Value

S3 object of class rescomp_event.

Examples

pars <- spec_rescomp(
  events = list(
    event_schedule_fixed(
      event = event_res_custom(function(res, params) {
        res * 2
      }),
      times = 500
    )
  )
)
plot_rescomp(sim_rescomp(pars))

Define an event in which the resource populations are multiplied by some factor

Description

Produces an event object a pulse of multiplicatively increased/decreased resource populations, e.g. a density-independent disturbance which kills prey species.

Usage

event_res_mult(resources_mult)

Arguments

resources_mult

A numeric vector or rescomp_coefs_vector by which to multiply the current resource concentrations. Should be greater than 1 for an increase, or less than 1 for a decrease. Should not be negative.

Value

S3 object of class rescomp_event.

Examples

m1 <- spec_rescomp(
  spnum = 2, resnum = 2,
  funcresp = funcresp_type1(crmatrix(
    0.10, 0.08,
    0.06, 0.12
  )),
  events = list(
    event_schedule_periodic(
      event_res_mult(c(0.5, 1.5)),
      period = 250
    )
  )
)
plot_rescomp(sim_rescomp(m1))

Define a schedule of an event occurring at a fixed set of times

Description

Produces an event schedule object suitable for passing in the events argument to spec_rescomp().

Usage

event_schedule_fixed(event_obj, times, priority = 0)

Arguments

event_obj

An object of class rescomp_event.
times

A numeric vector of times at which the event should occur.
priority

A number. If two events occur at the same time, the event with the lower priority number is processed first.

Value

S3 object of class rescomp_event_schedule.

Examples

m1 <- spec_rescomp(
  events = list(
    event_schedule_fixed(
      event_res_add(0.5),
      times = c(200, 400)
    ),
    event_schedule_fixed(
      event_sp_add(-500),
      times = c(600, 800)
    )
  )
)
plot_rescomp(sim_rescomp(m1))

Define a schedule of an event occurring repeatedly at a fixed period

Description

Produces an event schedule object suitable for passing in the events argument to spec_rescomp().

Usage

event_schedule_periodic(event_obj, period, start_time = period, priority = 0)

Arguments

event_obj

An object of class rescomp_event.
period

A number: the period at which the event occurs.
start_time

A number: the time at which the event first occurs.
priority

A number. If two events occur at the same time, the event with the lower priority number is processed first.

Value

S3 object of class rescomp_event_schedule.

Examples

m1 <- spec_rescomp(
  events = list(
    event_schedule_periodic(
      event_batch_transfer(dilution = 0.1, resources = 1),
      period = 250
    ),
    event_schedule_periodic(
      event_batch_transfer(dilution = 0.5, resources = 1),
      period = 250,
      start_time = 125
    )
  )
)
plot_rescomp(sim_rescomp(m1))

Define a schedule of one-off species introductions: one for each species.

Description

Produces an event schedule object suitable for passing in the events argument to spec_rescomp().

Usage

event_set_introseq(times, concentrations, priority = 0)

Arguments

times

A numeric vector, of length spnum, of times at which each species should be introduced.
concentrations

A numeric vector or rescomp_coefs_vector, of length spnum, of species concentrations; the concentration of each species that should be added at the corresponding time.
priority

A number. If two events occur at the same time, the event with the lower priority number is processed first.

Value

S3 object of class rescomp_event_schedule.

Examples

m1 <- spec_rescomp(
  spnum = 4, resnum = 4,
  funcresp = funcresp_type1(crmatrix(
    0.10, 0.05, 0.05, 0.05,
    0.05, 0.10, 0.05, 0.05,
    0.05, 0.05, 0.10, 0.05,
    0.05, 0.05, 0.05, 0.10
  )),
  events = list(
    event_schedule_periodic(
      event_batch_transfer(dilution = 0.5, resources = 1),
      period = 1000 / 16,
      priority = 0
    ),
    event_set_introseq(
      times = c(0, 250, 500, 750),
      concentrations = c(10, 100, 1000, 2000),
      priority = 1
    )
  ),
  cinit = 0
)
plot_rescomp(sim_rescomp(m1))

Define an event in which a constant amount of some consumers is added to (or removed from) the system

Description

Produces an event object representing a pulse of added/removed consumers, e.g. a species introduction.

Usage

event_sp_add(species, min_zero = TRUE)

Arguments

species

A numeric vector or rescomp_coefs_vector of species concentrations, by which the current species concentrations are increased. Can be negative, to decrease species concentrations.
min_zero

If this is TRUE, resulting species concentrations are clamped to a minimum of zero. If this is FALSE, negative values of species may reduce species concentrations below zero.

Value

S3 object of class rescomp_event.

Examples

m1 <- spec_rescomp(
  spnum = 2, resnum = 2,
  funcresp = funcresp_type1(crmatrix(
    0.10, 0.08,
    0.06, 0.12
  )),
  events = list(
    event_schedule_periodic(
      event_sp_add(c(-100, 100)),
      period = 250
    )
  )
)
plot_rescomp(sim_rescomp(m1))

Define an event which instantaneously changes species concentrations using an arbitrary function

Description

Produces an event object suitable for building event schedules to pass spec_rescomp().

Usage

event_sp_custom(func, spnum = NULL)

Arguments

func

A function that takes species (a numeric vector of species concentrations) and params (a list of parameters) and returns a numeric vector of species concentrations.
spnum

The number of species.

Details

If spnum is NULL, spec_rescomp() will attempt to infer it. This is fine if the result is used directly in an event schedule, but may fail if the result must be combined with other rescomp_events first.

Value

S3 object of class rescomp_event.

Examples

pars <- spec_rescomp(
  events = list(
    event_schedule_fixed(
      event = event_sp_custom(function(sp, params) {
        sp * 2
      }),
      times = 500
    )
  )
)
plot_rescomp(sim_rescomp(pars))

Define an event in which the consumer populations are multiplied by some factor

Description

Produces an event object a pulse of multiplicatively increased/decreased consumer populations, e.g. a density-independent disturbance which kills consumer species.

Usage

event_sp_mult(species_mult)

Arguments

species_mult

A numeric vector or rescomp_coefs_vector by which to multiply the current species concentrations. Should be greater than 1 for an increase, or less than 1 for a decrease. Should not be negative.

Value

S3 object of class rescomp_event.

Examples

m1 <- spec_rescomp(
  spnum = 2, resnum = 2,
  funcresp = funcresp_type1(crmatrix(
    0.10, 0.08,
    0.06, 0.12
  )),
  events = list(
    event_schedule_periodic(
      event_sp_mult(c(0.5, 1.5)),
      period = 250
    )
  )
)
plot_rescomp(sim_rescomp(m1))

Convert object of class deSolve to a data frame and name columns

Description

Convert object of class deSolve to a data frame and name columns

Usage

frame_and_name(model)

Arguments

model

List output from sim_rescomp(). First element is an object of class deSolve. Second element is an S3 object of class rescomp.

Value

data frame

Examples

pars <- spec_rescomp()
m1 <- sim_rescomp(pars)
frame_and_name(m1)

Define a functional response using an arbitrary function

Description

Produces an object suitable to pass as the funcresp to spec_rescomp().

Usage

funcresp_custom(func, spnum = NULL, resnum = NULL)

Arguments

func

A function that takes resources (a numeric vector of resource concentrations) and params (a list of parameters) and returns a matrix of growth rates of each species on each resource.
spnum

The number of species; the number of rows in the matrix returned by func.
resnum

The number of resources; the expected length of the resources argument to func and the number of columns in the matrix returned by func.

Details

If spnum or resnum are NULL, spec_rescomp() will attempt to infer them. This is fine if the result is passed directly to spec_rescomp(), but may fail if the result must be combined with other rescomp_funcresps first.

Value

S3 object of class rescomp_funcresp.

Examples

# Type 1 functional response with fixed growth rates
m1 <- spec_rescomp(
  spnum = 2, resnum = 3,
  funcresp_custom(
    function(resources, params) {
      growth_rates <- c(0.2, 0.3)
      outer(growth_rates, resources)
    },
    spnum = 2
  )
)
plot_funcresp(m1)

Define a Hill functional response

Description

Produces an object suitable to pass as the funcresp to spec_rescomp(). Creates a Hill functional response with maximum growth rate mumax, half saturation constant ks, and Hill coefficient n. mu_ij(R_j) = mumax_ij * (R_j)^n_ij / (ks_ij + (R_j)^n_ij) This an alternative parameterisation of a type 3 functional response, typically used with quota rather than efficiency.

Usage

funcresp_hill(mumax, ks, n)

Arguments

mumax

A matrix or rescomp_coefs_matrix, with one row per species and one column per resource. The maximum growth rate (if using quota) or maximum consumption rate (if using efficiency) of each species on each resource.
ks

A matrix or rescomp_coefs_matrix, with one row per species and one column per resource. The half saturation constant of each species for each resource; the resource concentration at which growth/consumption rate is half of mumax.
n

A matrix or rescomp_coefs_matrix, with one row per species and one column per resource. The Hill coefficient for each species on each resource. With n = 1, this is equivalent to a Monod functional response.

Value

S3 object of class rescomp_funcresp.

Examples

m1 <- spec_rescomp(
  spnum = 3,
  funcresp = funcresp_hill(
    mumax = crmatrix(0.2, 0.4, 0.4),
    ks = crmatrix(0.2, 1, 1),
    n = crmatrix(1, 1, 2)
  )
)
plot_funcresp(m1, maxx = 3)

Define a Monod functional response

Description

Produces an object suitable to pass as the funcresp to spec_rescomp(). Creates a Monod functional response with maximum growth rate mumax and half saturation constant ks. mu_ij(R_j) = mumax_ij * R_j / (ks_ij + R_j) This an alternative parameterisation of a type 2 functional response, typically used with quota rather than efficiency.

Usage

funcresp_monod(mumax, ks)

Arguments

mumax

A matrix or rescomp_coefs_matrix, with one row per species and one column per resource. The maximum growth rate (if using quota) or maximum consumption rate (if using efficiency) of each species on each resource.
ks

A matrix or rescomp_coefs_matrix, with one row per species and one column per resource. The half saturation constant of each species for each resource; the resource concentration at which growth/consumption rate is half of mumax.

Value

S3 object of class rescomp_funcresp.

Examples

m1 <- spec_rescomp(
  spnum = 2,
  funcresp = funcresp_monod(
    mumax = crmatrix(0.2, 0.4),
    ks = crmatrix(0.2, 1)
  )
)
plot_funcresp(m1, maxx = 3)

Define a linear functional response

Description

Produces an object suitable to pass as the funcresp to spec_rescomp(). Creates a linear or type 1 functional response with attack rate a. mu_ij(R_j) = a_ij * R_j

Usage

funcresp_type1(a)

Arguments

a

A matrix or rescomp_coefs_matrix, with one row per species and one column per resource. The attack rate (if using efficiency) or growth rate (if using quota) of each species on each resource.

Value

S3 object of class rescomp_funcresp.

Examples

m1 <- spec_rescomp(
  spnum = 2, resnum = 2,
  funcresp = funcresp_type1(
    crmatrix(
      0.2, 0.3,
      0.4, 0.2
    )
  )
)
plot_funcresp(m1)

m2 <- spec_rescomp(
  spnum = 2, resnum = 2,
  funcresp = funcresp_type1(
    rescomp_coefs_lerp(
      crmatrix(
        0.2, 0.3,
        0.4, 0.2
      ),
      crmatrix(
        0.2, 0.3,
        0.1, 0.1
      ),
      "antibiotic_concentration"
    )
  ),
  params = list(antibiotic_concentration = rescomp_param_square())
)
plot_funcresp(m2, display_values = list(antibiotic_concentration = c(0, 0.5, 1)))

Define a Holling type 2 functional response

Description

Produces an object suitable to pass as the funcresp to spec_rescomp(). Creates a Holling type 2 functional response with attack rate a and handling time h. mu_ij(R_j) = (a_ij * R_j) / (1 + a_ij * h_ij * R_j)

Usage

funcresp_type2(a, h)

Arguments

a

A matrix or rescomp_coefs_matrix, with one row per species and one column per resource. The maximum attack rate (if using efficiency) or growth rate (if using quota) of each species on each resource, when the resource is rare.
h

A matrix or rescomp_coefs_matrix, with one row per species and one column per resource. The handling time of each species for each resource, which reduces attack/growth rate when the resource becomes abundant. With h = 0, this is equivalent to a type 1 function response.

Value

S3 object of class rescomp_funcresp.

Examples

m1 <- spec_rescomp(
  spnum = 2,
  funcresp = funcresp_type2(
    a = crmatrix(0.4, 0.2),
    h = crmatrix(3, 1)
  )
)
plot_funcresp(m1, maxx = 3)

Define a Holling type 3 functional response

Description

Produces an object suitable to pass as the funcresp to spec_rescomp(). Creates a Holling type 3 functional response with attack rate a, handling time h, and exponent k. mu_ij(R_j) = (a_ij * (R_j)^k_ij) / (1 + a_ij * h_ij * (R_j)^k_ij)

Usage

funcresp_type3(a, h, k)

Arguments

a

A matrix or rescomp_coefs_matrix, with one row per species and one column per resource. The maximum attack rate (if using efficiency) or growth rate (if using quota) of each species on each resource, when the resource is rare.
h

A matrix or rescomp_coefs_matrix, with one row per species and one column per resource. The handling time of each species for each resource, which reduces attack/growth rate when the resource becomes abundant.
k

A matrix or rescomp_coefs_matrix, with one row per species and one column per resource. The exponent (a.k.a. Hill coefficient) for each species on each resource. With k = 1, this is equivalent to a type 2 functional response. Each k must be greater than 1 for a true type 3 functional response.

Value

S3 object of class rescomp_funcresp.

Examples

m1 <- spec_rescomp(
  spnum = 3,
  funcresp = funcresp_type3(
    a = crmatrix(0.5, 0.3, 0.2),
    h = crmatrix(4, 2, 1),
    k = crmatrix(2)
  )
)
plot_funcresp(m1, maxx = 3)

Get coefficients from a vector/matrix or rescomp_coefs_vector/rescomp_coefs_matrix object

Description

Provides a generic interface to a vector/matrix or a rescomp_coefs_vector/rescomp_coefs_matrix object. Called on a numeric vector/matrix (as appropriate to the function used), this will just return the vector/matrix. Called on a rescomp_coefs_vector/rescomp_coefs_matrix, this allows time-dependence via parameters. get_coefs() works on vector- or matrix-type objects, while the others work only on their specific types.

Usage

get_coefs(coefs_obj, params)

get_coefs_vector(coefs_obj, params)

get_coefs_matrix(coefs_obj, params)

Arguments

coefs_obj

A numeric vector/matrix or an object of class rescomp_coefs_vector/rescomp_coefs_matrix.
params

A list of time-dependent parameters.

Details

This function is normally only for internal use, but is exported to aid users in debugging their created rescomp_coefs objects.

Value

A vector/matrix of coefficients.

Examples

get_coefs(c(0.2, 0.3, 0.4), list())
get_coefs(matrix(c(0.2, 0.3, 0.4, 0.2), nrow = 2), list())
get_coefs_vector(c(0.2, 0.3, 0.4), list())
try(get_coefs_matrix(c(0.2, 0.3, 0.4), list()))

coefs <- rescomp_coefs_matrix_custom(
  function(params) {
    matrix(c(0.2, 0.3, 0.4, params$fourth_coeff), nrow = 2, ncol = 2)
  },
  nrow = 2, ncol = 2
)
get_coefs(coefs, list(fourth_coeff = 0.5))

Gets the list of times from a rescomp_event_schedule object

Description

This function is normally only for internal use, but is exported to aid users in debugging their created rescomp_event_schedule objects.

Usage

get_event_times(event_schedule_obj, totaltime)

Arguments

event_schedule_obj

An object of class rescomp_event_schedule.
totaltime

The total time which the simulation will run for.

Value

A vector of times.

Examples

schedule1 <- event_schedule_periodic(
  event_batch_transfer(dilution = 0.1, resources = 1),
  period = 250
)
get_event_times(schedule1, 1000)
get_event_times(schedule1, 999)

schedule2 <- event_schedule_periodic(
  event_batch_transfer(dilution = 0.1, resources = 1),
  period = 250,
  start_time = 125
)
get_event_times(schedule2, 1000)
get_event_times(schedule2, 999)

Get growth rates from a rescomp_funcresp object

Description

Gets the growth rates of each species on each resource, given resource concentrations. These must be combined, according to whether the resouces are essential or substitutable, to get the overall growth rate for each species.

Usage

get_funcresp(funcresp_obj, spnum, resources, params)

Arguments

funcresp_obj

An object of class rescomp_funcresp.
spnum

The number of species.
resources

A vector of resource concentrations.
params

A list of time-dependent parameters.

Details

This function is normally only for internal use, but is exported to aid users in debugging their created rescomp_funcresp objects.

Value

A matrix of species growth rates on each resource, with spnum rows and length(resources) columns.

Examples

funcresp <- funcresp_custom(
  function(resources, params) {
    growth_rates <- params$scale * c(0.2, 0.3)
    outer(growth_rates, resources)
  },
  spnum = 2
)
get_funcresp(funcresp, 2, c(1, 4, 5, 6), list(scale = 2))
get_funcresp(funcresp, 2, 0.7, list(scale = 0.5))
try(get_funcresp(funcresp, 3, 0.7, list(scale = 0.5)))

Get params at an instant in time from a rescomp_param object

Description

This function is normally only for internal use, but is exported to aid users in debugging their created rescomp_param objects.

Usage

get_params(param_obj, t)

Arguments

param_obj

An object of class rescomp_param.
t

The time at which to get the parameters.

Value

A numeric vector of length 1; the parameter value.

Examples

# Repeated pulsing and exponential decay
antibiotic_conc <- rescomp_param_custom(function(t) {
  0.5^(10 * (t %% 1))
})
get_params(antibiotic_conc, 0)
get_params(antibiotic_conc, 0.5)
get_params(antibiotic_conc, 1)

params <- list(r = 0.2, antibiotic_conc = antibiotic_conc)
get_params(params, 0.5)

Get resource supply rates from a rescomp_ressupply object

Description

Gets the resource supply rates of each resource, given the current resource concentrations.

Usage

get_ressupply(ressupply_obj, resources, params)

Arguments

ressupply_obj

An object of class rescomp_funcresp.
resources

A vector of resource concentrations.
params

A list of time-dependent parameters.

Details

This function is normally only for internal use, but is exported to aid users in debugging their created rescomp_ressupply objects.

Value

A vector of rates of change of resource concentrations, of the same length as resources.

Examples

# Two resources, A and B, with constant supply of A, and A spontaneously converting to B
ressupply <- ressupply_custom(
  function(resources, params) {
    conversion <- params$conversion * resources[1]
    return(c(params$supply - conversion, conversion))
  },
  resnum = 2
)
get_ressupply(ressupply, c(10, 20), list(supply = 3, conversion = 0.2))
try(get_ressupply(ressupply, c(10, 20, 30), list(supply = 3, conversion = 0.2)))

Plot functional responses

Description

Plot functional responses

Usage

plot_funcresp(pars, maxx = 1, display_values, madj = FALSE)

Arguments

pars

S3 object of class rescomp detailing model parameters and specifications.
maxx

Numeric vector of length 1. Resource value to calculate per-capita growth rates up to (xlim).
display_values

Named list of vectors, with names matching names of pars$params. Each vector gives the values of the respective model parameters at which to plot the functional responses. Defaults are automatically inferred for most rescomp_param objects, but may be overwritten. rescomp_param_custom parameters must have their display values specified here if they were not defined at definition.
madj

Logical vector of length 1. Whether to standardise per capita growth rates by mortality.

Details

It is assumed that the funcresp of pars is constructed such that the growth rate of a species on a given resource depends only on the concentration of that resource, and not on other resources. This is the case for all built-in functional responses, but is not necessary the case if using funcresp_custom(). Plots are likely to be nonsensical or incorrect if this assumption is violated.

Value

A ggplot object.

Examples

pars <- spec_rescomp()
plot_funcresp(pars)

pars <- spec_rescomp(
  spnum = 2,
  resnum = 2,
  funcresp = funcresp_monod(
    mumax = crmatrix(
      0.7, 0.3,
      0.4, 0.5
    ),
    ks = crmatrix(1)
  )
)
plot_funcresp(pars)
plot_funcresp(pars, madj = TRUE)
plot_funcresp(pars, maxx = 5)

pars <- spec_rescomp(
  spnum = 2,
  funcresp = funcresp_type1(
    a = rescomp_coefs_lerp(
      crmatrix(0.12, 0.08),
      crmatrix(0.08, 0.12),
      param_name = "temperature"
    )
  ),
  params = list(temperature = rescomp_param_sine(period = 250))
)
plot_funcresp(pars)
plot_funcresp(pars, display_values = list(temperature = c(0.0, 0.25, 0.5, 0.75, 1.0)))

Plot consumer and resource dynamics from deSolve output

Description

Plot consumer and resource dynamics from deSolve output

Usage

plot_rescomp(
  model,
  consumers = TRUE,
  resources = TRUE,
  logy = FALSE,
  lwd = 1,
  xlims = NULL
)

Arguments

model

List output from sim_rescomp(). First element is an object of class deSolve. Second element is an object of class rescomp.
consumers

Plot consumer dynamics? Default = TRUE.
resources

Plot resource dynamics? Default = TRUE.
logy

Log transfrom y-axis (default = FALSE).
lwd

Line width (default = 1)
xlims

Vector of length giving the time frame to plot.

Value

ggplot object

Examples

pars <- spec_rescomp()
m1 <- sim_rescomp(pars)
plot_rescomp(m1)

Create a set of coefficients by linear interpolation

Description

Produces an object that may be used in place of a vector or matrix in the creation of arguments to spec_rescomp(). Linearly interpolates between two sets of coefficients according to the value of a time-dependent parameter.

Usage

rescomp_coefs_lerp(coefs0, coefs1, param_name, param0 = 0, param1 = 1)

Arguments

coefs0, coefs1

Vectors, matrices, or objects of class rescomp_coefs_vector or rescomp_coefs_matrix to be interpolated between. Either both must be vectors, or both must be matrices.
param_name

A character vector of length 1; the name of the parameter to be interpolated according to.
param0, param1

The values of the parameter to take as the fixed points for interpolation.

Details

Coefficients are taken to have the values in coef0 at param = param0, and the values in coef1 at param = param1, and are linearly interpolated between. Each parameter in the vector/matrix is interpolated independently.

Value

S3 object of class rescomp_coefs_vector or rescomp_coefs_matrix, according to the types of coefs0 and coefs1.

Examples

coefs_vec <- rescomp_coefs_lerp(
  c(0, 0, 0),
  c(2, 3, 5),
  "scale"
)
get_coefs(coefs_vec, list(scale = 0))
get_coefs(coefs_vec, list(scale = 0.5))
get_coefs(coefs_vec, list(scale = 1))
get_coefs(coefs_vec, list(scale = 1.5))

coefs_mat1 <- rescomp_coefs_lerp(
  matrix(c(0.2, 0.4, 0.3, 0.2), nrow = 2),
  matrix(c(0.2, 0.1, 0.3, 0.1), nrow = 2),
  "antibiotic1_concentration",
  param0 = 0,
  param1 = 200
)
get_coefs(coefs_mat1, list(antibiotic1_concentration = 0))
get_coefs(coefs_mat1, list(antibiotic1_concentration = 100))
get_coefs(coefs_mat1, list(antibiotic1_concentration = 200))

coefs_mat2 <- rescomp_coefs_lerp(
  coefs_mat1,
  matrix(c(0.0, 0.0, 0.0, 0.0), nrow = 2),
  "antibiotic2_concentration",
  param0 = 0,
  param1 = 100
)
get_coefs(coefs_mat2, list(antibiotic1_concentration = 100, antibiotic2_concentration = 50))

Create a matrix of coefficients using an arbitrary function

Description

Produces an object that may be used in place of a matrix in the creation of arguments to spec_rescomp().

Usage

rescomp_coefs_matrix_custom(func, nrow, ncol)

Arguments

func

A function that takes params (a list of parameters) and returns a matrix.
nrow

The number of rows in the matrix returned by func.
ncol

The number of columns in the matrix returned by func.

Value

S3 object of class rescomp_coefs_matrix.

Examples

# Matrix with one time-varying coefficient
rescomp_coefs_matrix_custom(
  function(params) {
    matrix(c(0.2, 0.3, 0.4, params$fourth_coeff), nrow = 2, ncol = 2)
  },
  nrow = 2, ncol = 2
)

Create a vector of coefficients using an arbitrary function

Description

Produces an object that may be used in place of a vector in the creation of arguments to spec_rescomp().

Usage

rescomp_coefs_vector_custom(func, length)

Arguments

func

A function that takes params (a list of parameters) and returns a vector.
length

The length of the vector returned by func.

Value

S3 object of class rescomp_coefs_vector.

Examples

# Vector with one time-varying coefficient
rescomp_coefs_vector_custom(
  function(params) {
    c(0.2, 0.3, 0.4, params$fourth_coeff)
  },
  length = 4
)

Create a rescomp parameter using an arbitrary function

Description

Produces an object suitable to include in a list of rescomp parameters, providing time-dependence.

Usage

rescomp_param_custom(func, display_values = NULL)

Arguments

func

A function that takes t (time) and returns a number to use as a parameter.
display_values

A numeric vector of values of the parameter to use in plot_funcresp(). If not specified here, must be specified in plot_funcresp().

Value

S3 object of class rescomp_param.

Examples

# Repeated pulsing and exponential decay
antibiotic_conc <- rescomp_param_custom(function(t) {
  0.5^(10 * (t %% 1))
})
get_params(antibiotic_conc, 0)
get_params(antibiotic_conc, 0.5)
get_params(antibiotic_conc, 1)

Create a rescomp parameter using a sine/square/triangle wave

Description

Produces an object suitable to include in a list of rescomp parameters, providing time-dependence. Triangle and square waves are phase-shifted to be similiar in shape to a sine wave with the same period and offset, such that the peaks and troughs occur in the same places.

Usage

rescomp_param_sine(
  period = 1,
  min = 0,
  max = 1,
  offset = 0,
  display_values = c(min, max)
)

rescomp_param_triangle(
  period = 1,
  min = 0,
  max = 1,
  offset = 0,
  display_values = c(min, max)
)

rescomp_param_square(
  period = 1,
  min = 0,
  max = 1,
  offset = 0,
  display_values = c(min, max)
)

Arguments

period

The period of the wave.
min

The minimum value of the parameter; the mean minus the amplitude.
max

The maximum value of the parameter; the mean plus the amplitude.
offset

The phase shift of the wave. For a sine or triangle wave the time at which its value is equal to the mean and increasing. For a square wave, the time at which it increases to the maximum value.
display_values

A numeric vector of values of the parameter to use in plot_funcresp().

Value

S3 object of class rescomp_param.

Examples

sine <- rescomp_param_sine(period = 1)
cosine <- rescomp_param_sine(period = 1, offset = -0.25)
triangle <- rescomp_param_triangle(period = 1)
square <- rescomp_param_square(period = 1)

times <- seq(from = 0, to = 2, by = 0.01)
plot(times, get_params(sine, times), type = "l", col = "black")
lines(times, get_params(cosine, times), col = "blue")
lines(times, get_params(triangle, times), col = "maroon3")
lines(times, get_params(square, times), col = "orange")

Create a resource supply rate using chemostat dynamics

Description

Produces an object suitable to pass as the ressupply to spec_rescomp().

Usage

ressupply_chemostat(dilution, concentration)

Arguments

dilution

A numeric vector or rescomp_coefs_vector, of length one.
concentration

A vector or rescomp_coefs_vector, with one number per resource. The concentration of each resource in the incoming medium.

Value

S3 object of class rescomp_ressupply.

Examples

ressupply <- ressupply_chemostat(
  dilution = 0.01,
  concentration = rescomp_coefs_lerp(c(0, 0, 0), c(2, 3, 4), "ressupply_scaling")
)
get_ressupply(ressupply, c(2, 4, 10), list(ressupply_scaling = 0))
get_ressupply(ressupply, c(3, 3, 3), list(ressupply_scaling = 1))

Create a resource supply rate using a constant rate resource supply

Description

Produces an object suitable to pass as the ressupply to spec_rescomp().

Usage

ressupply_constant(rate)

Arguments

rate

A vector or rescomp_coefs_vector, with one number per resource. The supply rate of each resource.

Value

S3 object of class rescomp_ressupply.

Examples

ressupply <- ressupply_constant(c(0.2, 0.3))
get_ressupply(ressupply, c(2, 10), list())
get_ressupply(ressupply, c(5, 20), list())
# The above two give the same result; constant supply doesn't depend on existing concentration.

ressupply <- ressupply_constant(rescomp_coefs_lerp(c(0.2, 0.3), c(0.4, 0.6), "extra_supply"))
get_ressupply(ressupply, c(2, 10), list(extra_supply = 0.2))
get_ressupply(ressupply, c(2, 10), list(extra_supply = 0.8))

Create a resource supply rate using an arbitrary function

Description

Produces an object suitable to pass as the ressupply to spec_rescomp().

Usage

ressupply_custom(func, resnum = NULL)

Arguments

func

A function that takes resources (a numeric vector of resource concentrations) and params (a list of parameters) and returns a vector of rates of change of each resource.
resnum

The number of resources; the expected length of the resources argument to func and the length of the vector returned by func.

Details

If resnum is NULL, spec_rescomp() will attempt to infer it. This is fine if the result is passed directly to spec_rescomp(), but may fail if the result must be combined with other rescomp_ressupply first.

Value

S3 object of class rescomp_ressupply.

Examples

# Two resources, A and B, with constant supply of A, and A spontaneously converting to B
ressupply <- ressupply_custom(
  function(resources, params) {
    conversion <- params$conversion * resources[1]
    return(c(params$supply - conversion, conversion))
  },
  resnum = 2
)
get_ressupply(ressupply, c(10, 20), list(supply = 3, conversion = 0.2))

Create a resource supply rate using logistic resource growth

Description

Produces an object suitable to pass as the ressupply to spec_rescomp().

Usage

ressupply_logistic(r, k)

Arguments

r

A vector or rescomp_coefs_vector, with one number per resource. The intrinsic growth rate of each resource.
k

A vector or rescomp_coefs_vector, with one number per resource. The carrying capacity of each resource.

Value

S3 object of class rescomp_ressupply.

Examples

ressupply <- ressupply_logistic(
  r = rescomp_coefs_lerp(c(0.2, 0.3), c(0, 0), "growth_inhibition"),
  k = c(10, 20)
)
get_ressupply(ressupply, c(2, 10), list(growth_inhibition = 0))
get_ressupply(ressupply, c(20, 0), list(growth_inhibition = 0))
get_ressupply(ressupply, c(2, 10), list(growth_inhibition = 0.8))

Simulate resource competition (a convenience wrapper for deSolve::ode())

Description

Simulate resource competition (a convenience wrapper for deSolve::ode())

Usage

sim_rescomp(pars, totaltime, cinit, rinit, ...)

Arguments

pars

S3 object of class rescomp returned by rescomp::spec_rescomp().
totaltime

Numeric vector of length 1: the total simulation time. If provided, overrides the value in pars.
cinit

Numeric vector of length 1 or length spnum specifying initial consumer state values (densities). If provided, overrides the value in pars.
rinit

Numeric vector of length 1 or length resnum specifying initial resource state values (concentrations). If provided, overrides the value in pars.
...

Other arguments passed to deSolve::ode()

Value

A list of two comprising i) the model dynamics and ii) model specifications.

Examples

pars <- spec_rescomp()
results1 <- sim_rescomp(pars = pars)
plot_rescomp(results1)

results2 <- sim_rescomp(pars = pars, totaltime = 100, cinit = 1000)
plot_rescomp(results2)

Generate list of parameters for a consumer-resource model to be passed to sim_rescomp()

Description

Generate list of parameters for a consumer-resource model to be passed to sim_rescomp()

Usage

spec_rescomp(
  spnum = 1,
  resnum = 1,
  funcresp = funcresp_type1(crmatrix(0.1)),
  quota = crmatrix(0.001),
  efficiency,
  essential = FALSE,
  mort = 0.03,
  ressupply = ressupply_chemostat(0.03, 1),
  params = list(),
  events = list(),
  totaltime = 1000,
  cinit = 10,
  rinit = 1,
  verbose = FALSE
)

Arguments

spnum

Integer vector of length 1: the number of consumers.
resnum

Integer vector of length 1: the number of resources.
funcresp

An object of class rescomp_funcresp specifying the functional response.
quota

Numeric matrix or rescomp_coefs_matrix, the elements of which give the resource quotas. The number of rows and columns should be equal to spnum and resnum respectively. Mutually exclusive with efficiency.
efficiency

Numeric matrix or rescomp_coefs_matrix, the elements of which give the efficiency of each consumer on each resource. The number of rows and columns should be equal to spnum and resnum respectively. Mutually exclusive with quota.
essential

Logical vector of length 1. If FALSE resources are substitutable.
mort

Numeric vector or rescomp_coefs_vector of length spnum, specifying density independent mortality rates.
ressupply

An object of class rescomp_ressupply specifying the resource supply.
params

A list specifying a set of parameters which may vary with time, on which other parameters of the model (e.g. funcresp, ressupply) may depend. rescomp_param objects in the list are parsed to allow time dependence, while other objects are passed directly.
events

A list of objects of class rescomp_event_schedule, specifying events that instantaneously change consumer or resource densities.
totaltime

Numeric vector of length 1: the total simulation time.
cinit

Numeric vector of length 1 or length spnum specifying initial consumer state values (densities).
rinit

Numeric vector of length 1 or length resnum specifying initial resource state values (concentrations).
verbose

If TRUE (default) prints model and simulation summary specs.

Details

Only one of efficiency and quota should be specified. Specifying both is an error. The default, if neither is specified, is to use quota. If using quota, the functional responses are taken to give per capita growth rates. If using efficiency, the functional responses are taken to give attack rates.

Value

S3 object of class rescomp.

Examples

# Using default parameters.
m1 <- spec_rescomp()
plot_rescomp(sim_rescomp(m1))

# With two species, two resources, and type 2 functional responses.
m2 <- spec_rescomp(
  spnum = 2,
  resnum = 2,
  funcresp = funcresp_type2(
    a = crmatrix(
      0.1, 0.2,
      0.15, 0.15
    ),
    h = crmatrix(1)
  ),
  totaltime = 500
)
plot_rescomp(sim_rescomp(m2))

# With serial dilution (batch transfer) events.
m3 <- spec_rescomp(
  ressupply = ressupply_constant(0),
  mort = 0,
  events = list(
    event_schedule_periodic(
      event_batch_transfer(dilution = 0.1, resources = 1),
      period = 125
    )
  ),
  totaltime = 500
)
plot_rescomp(sim_rescomp(m3))

# Growth rates affected by seasonal temperature fluctuations.
m4 <- spec_rescomp(
  spnum = 2,
  funcresp = funcresp_type1(
    a = rescomp_coefs_lerp(
      crmatrix(0.12, 0.08),
      crmatrix(0.08, 0.12),
      param_name = "temperature"
    )
  ),
  params = list(temperature = rescomp_param_sine(period = 125)),
  totaltime = 500
)
plot_rescomp(sim_rescomp(m4))