geneDeletion: Gene Deletion Experiments
In sybil: Efficient Constrained Based Modelling

Description Usage Arguments Details Value Author(s) See Also Examples

The function geneDeletion studies the effect of n in silico gene deletions on the phenotype of a metabolic network. The value of n is the number of genes knocked-out simultaneously.

1 2	geneDeletion(model, genes, combinations = 1, lb = NULL, ub = NULL, checkOptSolObj = FALSE, ...)

`model`	An object of class `modelorg`.
`genes`	Character or Integer: the genes to delete (see Details below).
`combinations`	A single integer value. If `combinations > 1` and `genes` is not a matrix, `combinations` is the number of elements from `genes` taken at a time while building all combinations of the elements in `genes` (see Details below). Default: `1`.
`lb`	A numeric vector containing the lower bounds for the reaction rates of reactions (variables) affected by the genes given in argument `genes`. If set to `NULL`, all reactions affected will be constrained to zero. Default: `NULL`.
`ub`	A numeric vector containing the upper bounds for the reaction rates of reactions (variables) affected by the genes given in argument `genes`. If set to `NULL`, all reactions affected will be constrained to zero. Default: `NULL`.
`checkOptSolObj`	A single logical value. If set to `TRUE`, a warning will be generated, if not all optimizations ended successful. Default: `FALSE`.
`...`	Further arguments passed to `optimizer`. Important ones are `algorithm` in order to set the algorithm to use or `solverParm` in order to set parameter values for the optimization software.

If argument genes is a matrix of character values (gene id's) or integers (pointers to gene id's), each column is treated as one deletion experiment. If the matrix is made up of integers, a zero entry means no gene.

If argument genes is a character vector or integer, the argument combinations gives the number of gene id's taken each time in order to build all possible combinations of genes. A matrix is constructed using combn. The value of argument combinations gives the number of genes, which are knocked-out simultaneously. The default value 1 performs a single gene deletion experiment, like the function oneGeneDel does. A value of 2 performs a double gene deletion as described in doubleGeneDel. A value of n performs an n gene deletion experiment. Keep in mind, that the number of optimizations will get very high for increasing values of combinations.

If argument genes is empty, the number of unique genes present in model is used.

The required length of arguments lb and ub (if not NULL) depends on the values given in arguments genes and combinations. If genes is a matrix, lb and ub must be of length equal to the number of columns in genes. If genes is a vector, lb and ub must be of length equal to length(genes) * combinations.

An object of class optsol_genedel.

Gabriel Gelius-Dietrich <geliudie@uni-duesseldorf.de>

Maintainer: Mayo Roettger <mayo.roettger@hhu.de>

modelorg, optsol, optsol_genedel, checkOptSol, oneGeneDel, optimizer, optimizeProb, combn and SYBIL_SETTINGS.

  ## load the dataset
  data(Ec_core)
  
  ## perform a single gene deletion analysis
  ## (delete every gene one by one) via FBA
  gd <- geneDeletion(Ec_core)
  
  ## or via MOMA (linearized version)
  gd <- geneDeletion(Ec_core, algorithm = "lmoma")
  
  ## triple gene deletion analysis using the first ten genes
  gd <- geneDeletion(Ec_core, genes = 10, combinations = 3)

## Not run: 
  ## perform a double gene deletion analysis
  ##(delete all possible pairwise combinations of all genes)
  gd <- geneDeletion(Ec_core, combinations = 2)

  ## perform a triple gene deletion analysis
  ## (very high number of optimizations)
  gd <- geneDeletion(Ec_core, combinations = 3)

## End(Not run)