simul.evol.enz.multiple: Multiple simulations of enzyme evolution
In SimEvolEnzCons: Simulation of Evolution of Enzyme Under Constraints

Description Usage Arguments Details Value See Also Examples

View source: R/simul.evol.enz.multiple.R

This function gives multiple simulations of evolution of enzyme concentrations under constraints

simul.evol.enz.multiple(E_ini_fun, kin_fun, Keq_fun, nsim, 
N_fun, correl_fun, beta_fun=NULL, X_fun=1, pasobs=250, npt=500, 
same.E0=TRUE, is.random.E0=FALSE, same.kin0=TRUE, is.random.kin0=FALSE,
Etot_fun=100, kin_max=1000, max_mut_size_E=1, max_mut_size_A=1,
pmutA=0, typ_E=1, typ_A=1, use.old.mut=FALSE)

`E_ini_fun`	Numeric vector of the initial concentrations
`kin_fun`	Numeric vector of the initial kinetic parameters
`Keq_fun`	Numeric vector of equilibrium constants
`nsim`	Numeric. Number of simulations
`N_fun`	Numeric. Population size
`correl_fun`	Character string indicating the abbreviation of the constraint applied on the system
`beta_fun`	Matrix of co-regulation coefficients
`X_fun`	Numeric. Numerator of function `flux`. Default is 1
`pasobs`	Numeric. Number of time steps between two successive observations of the system. Default is `250`
`npt`	Numeric. Number of observations. Default is `500`
`same.E0, same.kin0`	Logical. Uses same `E_ini_fun` / `kin_fun` for all simulations ? Default is `TRUE`.
`is.random.E0, is.random.kin0`	Logical. Is `E_ini_fun` / `kin_fun` chosen randomly ? Default is `FALSE`.
`Etot_fun`	Numeric. Total concentration if `is.random.E0` is `TRUE`. Default is `100`.
`kin_max`	Numeric. Maximal value for random `kin_fun`. Default is `1000`.
`max_mut_size_E`	Numeric. Maximum absolute size of mutation for enzyme concentrations. Default is 1
`max_mut_size_A`	Numeric. Maximum absolute size of mutation for kinetic parameters. Default is 1
`pmutA`	Numeric. Mutation probability of kinetic parameters. Higher `pmutA`, higher the mutation probability of kinetic parameters compared to enzyme concentrations. Default is 0, i.e. no mutation of kinetic parameters
`typ_E`	Numeric for mutation method. Authorized values: 1 or 2. Default is 1.
`typ_A`	Numeric for mutation method. Default is 1. See details in `mut.kin`.
`use.old.mut`	Logical. If `FALSE` (default), use `mut.E.direct` mutation method, else use `mut.E.old` mutation method if `TRUE`

Details about how does simulations work are in function simul.evol.enz.one documentation.

Apply would also work, but only for multiple values of E_ini_fun. simul.evol.enz.multiple gives all results in a same table, and has the possibility to change kinetic parameters between simulation.

Initial parameters

You can choose if initial concentrations and initial kinetic parameters are identical between all simulations by setting same.E0 and same.kin0.

You can choose randomly initial concentrations and initial kinetic parameters by setting is.random.E0 and is.random.kin0.

Four cases are possible. Examples are given for initial concentrations, but are the same for kinetic parameters, where same.E0, is.random.E0 and E_ini_fun are respectively same.kin0, is.random.kin0 and kin_fun.

same.E0 is TRUE and is.random.E0 is FALSE is default parameters. In this case, input E_ini_fun of length n is used for all simulations.
If same.E0 and is.random.E0 are TRUE, E_ini_fun is chosen randomly and used for all nsim simulations.
If same.E0 is FALSE and is.random.E0 TRUE, each simulation has different random initial concentrations.
If same.E0 and is.random.E0 are FALSE, a matrix of n columns and nsim rows is needed for E_ini_fun, each row corresponding to one simulation.

$list_init is the return for initial concentrations, kinetic parameters and activities.

Other details

Note that n is the number of enzyme and is determined by length of Keq_fun.

Invisible list of 6 elements:

$tabR: dataframe of nsim x npt rows and 3*n+4 columns. Each row corresponds to state at each observation step (i.e. after pasobs mutations), and columns are respectively concentrations (1:n), kinetic parameters (n+1:2n), total concentration (2n+1), total kinetic (2n+2), flux/fitness (2n+3), activities (2n+4:3n+3), and simulation number ($sim or 3n+4);
$tabP_e: numeric matrix of npt rows and n+1 columns, corresponding to relative concentrations at equilibrium (column 1 to n) for each observation step (in rows), plus column $sim;
$tabP_r: same as $tabP_e, but for response coefficients
$list_init: list of 3 elements, containing initial values of concentrations in $E0, kinetic parameters in $kin0 and activities in $A0 for each simulation. Each element is a numeric matrix of nsim rows (one by simulation) and n columns (one by enzyme);
$list_final: list of 3 elements, containing final values of concentrations in $E_f, kinetic parameters in $kin_f and activities in $A_f for each simulation. Each element is a numeric matrix of nsim rows and n columns;
$param: list of input parameters.

Note that n is the number of enzymes, which is the length of E_ini_fun.

See function simul.evol.enz.one to see how works a simulation.

# case for 3 enzymes
n <- 3
E0 <- c(30,30,30)
kin <- c(1,10,30)
Keq <- c(1,1,1)
nsim <- 2 # 2 simulations
N <- 1000
beta <- diag(1,n)
beta[upper.tri(beta)] <- c(0.32,0.32*(-0.43),-0.43)
#beta_12 = 0.32, beta_13 = beta_12 x beta_23, beta_23 = -0.43
t_beta <- t(beta) #because R fills matrix column by column
beta[lower.tri(beta)] <- 1/t_beta[lower.tri(t_beta)] #beta_ji = 1/beta_ij
if (n==3) {beta[lower.tri(beta)] <- 1/beta[upper.tri(beta)]} #only available if n=3
correl <- "RegNeg"

evol_sim <- simul.evol.enz.multiple(E0,kin,Keq,nsim,N,correl,beta)
evol_sim$tabR[,1:n] #concentrations
evol_sim$tabR[,(n+1):(2*n)] #kinetic parameters
evol_sim$tabR[,(2*n+1)] #total concentration
evol_sim$tabR[,(2*n+2)] #total kinetic
evol_sim$tabR[,(2*n+3)] #flux
evol_sim$tabR[,(2*n+4):(3*n+3)] #activities
evol_sim$tabR$sim #simulation number
evol_sim$tabR[evol_sim$tabR$sim==2,] #results for 2nd simulation