FBA_solver {GCModeller} R Documentation

Flux Balance Analysis (FBA) Solver in Native R

Description

Solves a Flux Balance Analysis (FBA) problem by formulating it as a linear programming problem using the stoichiometric matrix and solving it with the \code{lpSolve} package. Results, including full and active flux distributions, are exported to the specified output directory.

Usage

FBA_solver(S.df, obj.rxns,
    outputdir = './',
    default.lb = -1000,
    default.ub = 1000,
    flux.bounds = 'list'());

Arguments

S.df

A stoichiometric matrix of the reaction network. It can be either a data.frame (where rows are metabolites and columns are reactions) or a file path to a CSV file. If a file path is provided, the first column is assumed to contain metabolite names (rownames) and the first row to contain reaction IDs (colnames).

obj.rxns

A character vector specifying the reaction flux IDs (a subset of the column names of \code{S_df}) to be maximized. If multiple reactions are provided, their combined flux is maximized.

outputdir

A character string specifying the directory path for exporting FBA result files. Defaults to the current directory (\code{"./"}).

default.lb

A numeric value specifying the default lower bound for every metabolic reaction flux. Defaults to \code{-1000} (assumes reversible reactions by default).

default.ub

A numeric value specifying the default upper bound for every metabolic reaction flux. Defaults to \code{1000}.

flux.bounds

A named list used to tweak specific reaction flux bounds. The name of each element should be the reaction ID, and the value should be a numeric vector of length 2 containing the lower and upper bounds: \code{list("flux_id" = c(lb, ub))}.

Details

The function formulates the standard FBA problem: \itemize{ \item Maximize: \code{sum(obj_rxns fluxes)} \item Subject to: \code{S * v = 0} (Steady-state constraint) \item And: \code{lb <= v <= ub} (Flux bounds) } If the solver succeeds, it writes the following files to \code{outputdir}: \itemize{ \item \code{FBA_Full_Flux_Distribution.csv}: All reactions and their calculated fluxes. \item \code{FBA_Active_Flux_Distribution.csv}: Reactions with flux absolute value > 1e-6. \item \code{ObjectiveFlux.txt}: The maximum objective value. \item \code{FBA_Result.json}: The raw result object from \code{lpSolve} in JSON format. } If the solver fails, it prints the status code and common troubleshooting tips to the console.

Authors

xieguigang

Value

Invisible \code{NULL}. The function is called for its side effects (console output and file generation).

Examples

#\dontrun{

# Assuming you have a stoichiometric matrix data frame 'S_mat'
# where colnames are reaction IDs and rownames are metabolites
# Define objective reactions
objectives <- c("BIOMASS_Ecoli_core_w_GAM")
# Define specific bounds for glucose uptake (e.g., max uptake = 10)
my_bounds <- list("EX_glc__D_e" = c(-10, 1000))
# Run FBA
FBA_solver(
S_df = S_mat,
obj_rxns = objectives,
outputdir = "./fba_results",
default_lb = -1000,
default_ub = 1000,
flux_bounds = my_bounds
)

#}
[Package GCModeller version 1.1.0-beta Index]