CODEWITHSUNDEEP
rfor-loopapplylapplyode
Javier
Javier

Reputation: 457

apply( )-like function retrieving information of several dataframes for substituting a loop

I have a problem that is as follows: I have two dataframes (parametersand amounts). 

> head(parameters)
        Ka       V1        V2   Q        CL
1 5.960224 20.46632  90.48670 5.5 68.192196
2 1.176143 16.05779 132.44959 5.5  6.421357
3 1.324476 34.91569  42.96544 5.5 20.761933
4 2.649055 21.55833 111.89387 5.5  6.868612
5 1.495101 32.97900 140.82672 5.5  4.340567
6 2.807148 29.33883 206.71401 5.5  8.756406

>head(amounts)
  Depot Central Peripheral
1    10       0          0
2    10       0          0
3    10       0          0
4    10       0          0
5    10       0          0
6    10       0          0

I would like to pass each row of parameters & amounts to a ODE-solver named ode, part of the deSolvepackage. I came with the following answer using a forloop, it works fine, but it takes ages when the number of loops is high (i.e, more than 500 iterations). I would like to substitute the forloop with an apply-like function. This is the loop:

Central_concs <- vector()

for(i in 1:nrow(parameters)){
  n_a <- names(amounts)
  amounts_i <- as.numeric(amounts[i, ])
  names(amounts_i) <- n_a
  parameters_i <- parameters[i, ]
  out <- ode(y = amounts_i, times = times, func = TwoComp, parms = parameters_i)
  Central_concs <- cbind(Central_concs, out)
}

The name extraction and re-naming is just an issue with the odefunction, don't worry about it. TwoComp, argument passed to ode, is just a function containing the system of differential equations.

I have tried with apply, but I have been totally unsuccessful. Any help is appreciated.

Upvotes: 0

Views: 37

Answers (2)

otwtm
otwtm

Reputation: 1999

Maybe something like this?

parameters_matrix <- as.matrix(parameters)
amoumts_matrix <- as.matrix(amounts)

myfun <- function(i){
  ode(y = amounts_matrix[i, ], times = times, func = TwoComp, parms = parameters_matrix[i,])
}


sapply(X =  1:nrow(parameters), FUN = myfun)

Upvotes: 3

desval
desval

Reputation: 2435

Maybe something along these lines:

custom.ode <- function(i){
  n_a <- names(amounts)
  amounts_i <- as.numeric(amounts[i, ])
  names(amounts_i) <- n_a
  parameters_i <- parameters[i, ]
  out <- ode(y = amounts_i, times = times, func = TwoComp, parms = parameters_i)
  return(out)
}


Central_concs <- lapply(1:nrow(amounts), custom.ode)

Upvotes: 2

Related Questions