R/model_to_pathway.R
In saezlab/ocean: metabolic enzyme enrichment analysis
Defines functions
nitrogen_tracking
split_transaminases
remove_cofactors
model_to_pathway_sif
Documented in
model_to_pathway_sif
nitrogen_tracking
remove_cofactors
split_transaminases
#'\code{model_to_pathway_sif}
#'
#' This function generate a subreaction network in sif format from the
#' recon_redhuman model for a given set of pathways
#'
#' @param pathway_to_keep character vector with the ID of the pathways that
#' should be used to build the SIF reaction network.
#' Use "View(unique(recon2_redhuman$pathway))" to see available pathway IDs
#' @param pathways pathway corresponding to redhuman reaction indexes
#' @param rxns reaction IDs corresponding to redhuman reaction indexes
#' @param rules gene rules corresponding to redhuman reaction indexes
#' @param minmax min max flux boundaries corresponding to redhuman reaction indexes
#' @param stochio redhuman stochiometric matrix
#' @param metab metabolite IDs corresponding to redhuman metabolite indexes
#' @param mapping obsolete
#' @param recon1_metabolites obsolete
#' @return a subreaction network in sif format
#' @export
model_to_pathway_sif <- function(pathway_to_keep,
pathways = recon2_redhuman$pathway, #redHuman_models folder
rxns = recon2_redhuman$rxns,
rules = recon2_redhuman$rules,
minmax = recon2_redhuman$minmax,
stochio = recon2_redhuman$stochio,
metab = recon2_redhuman$metab,
mapping = recon2_redhuman$gene_mapping,
recon1_metabolites = recon2_redhuman$metab_mapping){
####
index_to_keep <- sapply(pathway_to_keep, function(x, pathways){
which(pathways == x)
}, pathways = pathways, simplify = T)
index_to_keep <- unique(unlist(index_to_keep))
####
rxns <- rxns[index_to_keep,, drop = F]
rules <- rules[index_to_keep,, drop = F]
minmax <- minmax[index_to_keep,, drop = F]
pathways <- pathways[index_to_keep,, drop = F]
stochio <- stochio[,index_to_keep,drop = F]
####
print("Building gprs...")
gprs <- build_grps(rxns, rules, minmax, pathways)
####
print("Building reaction list...")
reactions_list <- build_reactions_list(gprs, stochio, metab, minmax)
###
print("Building reaction data frame...")
reactions_df <- build_reactions_df(reactions_list, mapping)
###
# print("Converting network gene IDs...")
# reactions_df <- translate_network_gene_ids(reactions_df, mapping)
reactions_df <- reactions_df[!grepl("HC0",reactions_df$V1) & !grepl("HC0",reactions_df$V2),]
reactions_df <- reactions_df[complete.cases(reactions_df),]
###
print("Converting network metabolite IDs...")
reactions_df <- translate_network_metab_ids(reactions_df, recon1_metabolites)
names(reactions_df) <- c("source","target")
reactions_df <- unique(reactions_df)
###
row.names(reactions_df) <- c(1:length(reactions_df[,1]))
if("Urea cycle" %in% pathway_to_keep)
{
print("Ataching Urea cycle additional reactions")
reactions_df <- as.data.frame(rbind(reactions_df,network_supplements[["Urea cycle"]]))
}
if("Purine synthesis" %in% pathway_to_keep)
{
print("Ataching Purine synthesis additional reactions")
reactions_df <- as.data.frame(rbind(reactions_df,network_supplements[["Purine synthesis"]]))
}
if("Valine, leucine, and isoleucine metabolism" %in% pathway_to_keep)
{
print("Ataching BCAA additional reactions")
reactions_df <- as.data.frame(rbind(reactions_df,network_supplements[["Carnitine synthesis"]])) #add ketoacid to carnitine
reactions_df <- as.data.frame(rbind(reactions_df,network_supplements[["Valine leucine and isoleucine metabolism"]])) #add bcat1 reverse
# reactions_df <- reactions_df[reactions_df[,1] != "1629_594_593_1738" & #remove the wrong BCKDH complex and correct it
# reactions_df[,2] != "1629_594_593_1738" &
# reactions_df[,1] != "1629_1738_594_593" &
# reactions_df[,2] != "1629_1738_594_593",]
reactions_df <- reactions_df[!grepl("CRAT.*_reverse",reactions_df[,1]) & #remove reverse crat
!grepl("CRAT.*_reverse",reactions_df[,2]),]
# reactions_df <- as.data.frame(rbind(reactions_df,network_supplements[["BCKDH"]])) #add correct BCKDH
}
###
return(reactions_df)
}
#'\code{remove_cofactors}
#'
#' This function removes cofactors from reaction networks generate by the model_to_pathway_sif function
#'
#' @param reaction_network reaction network in SIF format,
#' like the one returned by model_to_pathway_sif
#' @param compound_list KEGGREST coumpound list
#' @return reaction network in SIF format and the corresponding
#' attribute dataframe
#' @export
remove_cofactors <- function(reaction_network, compound_list = kegg_compounds)
{
compounds <- unique(c(reaction_network$source, reaction_network$target))
compounds <- compounds[grepl("cpd[:]",compounds)]
compounds <- as.data.frame(compounds)
names(compounds)[1] <- "compounds_compartment"
compounds$compound <- gsub("_.*","",compounds$compounds_compartment)
compound_vec <- unique(compounds$compound)
bad_kegg_compounds <- list()
for (compound in compound_list)
{
brite <- compound$BRITE
compound$NAME <- gsub(";","",compound$NAME)
if(!is.null(compound$ATOM))
{
brite <- gsub("[ ]+","",brite)
if(("Cofactors" %in% brite |
"Nucleotides" %in% brite |
"CO2" %in% compound$NAME |
as.numeric(compound$ATOM[1]) <= 3 |
"ITP" %in% compound$NAME |
"IDP" %in% compound$NAME |
"NADH" %in% compound$NAME |
"NADPH" %in% compound$NAME |
"FADH2" %in% compound$NAME) &
!"SAM" %in% compound$NAME)
{
bad_kegg_compounds[[compound$ENTRY[1]]] <- compound
}
}
}
bad_entries <- c()
for(compound in bad_kegg_compounds)
{
bad_entries <- c(bad_entries,compound$ENTRY[1])
}
bad_entries <- paste("cpd:",bad_entries,sep = "")
bad_compounds <- compounds[compounds$compound %in% bad_entries,]
##manually add diphosphate to be removed from network
bad_compounds <- as.data.frame(rbind(bad_compounds,c("cpd:C00013_c","cpd:C00013")))
bad_compounds <- as.data.frame(rbind(bad_compounds,c("cpd:C00013_m","cpd:C00013")))
bad_compounds <- as.data.frame(rbind(bad_compounds,c("cpd:C00009_c","cpd:C00009")))
bad_compounds <- as.data.frame(rbind(bad_compounds,c("cpd:C00009_m","cpd:C00009")))
bad_compounds <- as.data.frame(rbind(bad_compounds,c("cpd:C00009_r","cpd:C00009")))
reaction_network_no_cofact <- reaction_network[
!(reaction_network$source %in% bad_compounds$compounds_compartment) &
!(reaction_network$target %in% bad_compounds$compounds_compartment),
]
reaction_network_no_cofact <-
reaction_network_no_cofact[
complete.cases(reaction_network_no_cofact),
]
nodes <- unique(c(reaction_network_no_cofact$source ,
reaction_network_no_cofact$target ))
node_attributes <- as.data.frame(matrix(NA,length(nodes),2))
node_attributes[,1] <- nodes
node_attributes[,2] <- "reaction_enzyme"
node_attributes[node_attributes[,1] %in% compounds$compounds_compartment,2] <- "metabolite"
return(list("reaction_network" = reaction_network_no_cofact, "attributes" = node_attributes))
}
#'\code{compress_transporters}
#'
#' This function removes cofactors from reaction networks generate by the model_to_pathway_sif function
#'
#' @param sub_network_nocofact ipsum...
#' @return ipsum...
#' @export
compress_transporters <- function (sub_network_nocofact)
{
test_1 <- paste(gsub("_[cxrnme]$", "", sub_network_nocofact$reaction_network$source),
gsub("_[cxrnme]$", "", sub_network_nocofact$reaction_network$target),
sep = "_")
test_2 <- paste(gsub("_[cxrnme]$", "", sub_network_nocofact$reaction_network$target),
gsub("_[cxrnme]$", "", sub_network_nocofact$reaction_network$source),
sep = "_")
test_1[test_1 %in% test_2]
transporters <- unique(c(sub_network_nocofact$reaction_network[test_1 %in%
test_2, 1], sub_network_nocofact$reaction_network[test_1 %in%
test_2, 2]))
transporters <- transporters[!grepl("_[cxrnme]$", transporters)]
for (i in 1:2) {
sub_network_nocofact$reaction_network[, i] <- sapply(sub_network_nocofact$reaction_network[,
i], function(x, transporters) {
if (x %in% transporters) {
return("transporter")
}
else {
return(x)
}
}, transporters = transporters)
}
sub_network_nocofact$reaction_network$edgeId <- paste(sub_network_nocofact$reaction_network$source,
sub_network_nocofact$reaction_network$target, sep = "_")
sub_network_nocofact$reaction_network <- sub_network_nocofact$reaction_network[!duplicated(sub_network_nocofact$reaction_network$edgeId),
]
edge_transporter <- sub_network_nocofact$reaction_network$edgeId
edge_transporter <- edge_transporter[grepl("transporter",
edge_transporter)]
groups <- rep(0, length(edge_transporter))
metab_memory <- c()
for (i in 1:length(edge_transporter)) {
if (grepl("^cpd:", edge_transporter[i]) | grepl("_[cxrnme]_",
edge_transporter[i])) {
metab <- gsub("_.*", "", edge_transporter[i])
if(!metab %in% metab_memory)
{
metab_memory[i] <- metab
for (j in 1:length(edge_transporter)) {
if (grepl(metab, edge_transporter[j]))
{
groups[j] <- i
}
}
}
}
}
names(groups) <- edge_transporter
network <- sub_network_nocofact$reaction_network
for (i in 1:2) {
for (j in 1:length(network[, i])) {
if (network[j, i] == "transporter") {
print(network[j, 3])
network[j, i] <- paste(network[j, i], groups[network[j,
3]], sep = ">")
}
}
}
network <- network[!grepl("^SLC", network$edgeId) & !grepl("_SLC",
network$edgeId), ]
sub_network_nocofact$reaction_network <- network[, -3]
sub_network_nocofact$attributes <- sub_network_nocofact$attributes[sub_network_nocofact$attributes[, 1] %in% network[, 1] |
sub_network_nocofact$attributes[, 1] %in% network[, 2], ]
new_transporters <- unique(c(network[, 1], network[, 2]))
new_transporters <- new_transporters[grepl("transporter",
new_transporters)]
new_transporters <- as.data.frame(cbind(new_transporters,
rep("transporter", length(new_transporters))))
names(new_transporters) <- c("V1", "V2")
sub_network_nocofact$attributes <- as.data.frame(rbind(sub_network_nocofact$attributes,
new_transporters))
return(sub_network_nocofact)
}
#'\code{split_transaminases}
#'
#' This function removes cofactors from reaction networks generate by the model_to_pathway_sif function
#'
#' @param sub_network_nocofact ipsum...
#' @return ipsum...
#' @export
split_transaminases <- function(sub_network_nocofact)
{
sub_net <- sub_network_nocofact$reaction_network
transaminases_net <- sub_net[grepl("C0002[56]",sub_net$source) |
grepl("C0002[56]",sub_net$target) |
grepl("C00064",sub_net$source) |
grepl("C00064",sub_net$target),]
sub_net_no_transaminase <- sub_net[!(grepl("C0002[56]",sub_net$source) |
grepl("C0002[56]",sub_net$target) |
grepl("C00064",sub_net$source) |
grepl("C00064",sub_net$target)),]
transaminases_potential <- unique(c(transaminases_net$source,transaminases_net$target))
transaminases_potential <- transaminases_potential[!grepl("cpd:",transaminases_potential)]
transaminases_net <- sub_net[sub_net$source %in% transaminases_potential |
sub_net$target %in% transaminases_potential,]
splitted_transaminase <- sapply(transaminases_potential, function(transaminase, transaminases_net){
sub_net_transaminase <- transaminases_net[transaminase == transaminases_net$source |
transaminase == transaminases_net$target,]
columns <- c(1,2)
elements <- unique(c(sub_net_transaminase[,1],sub_net_transaminase[,2]))
if(sum(grepl("cpd:C00025_.",elements)) & sum(grepl("cpd:C00026_.",elements)))
{
for(i in 1:length(sub_net_transaminase[,1]))
{
for(j in 1:2)
{
if(grepl("C0002[56]",sub_net_transaminase[i,j]))
{
sub_net_transaminase[i,columns[-j]] <- paste(sub_net_transaminase[i,columns[-j]],"_gluakg",sep = "")
}
}
}
}
if(sum(grepl("cpd:C00025_.",elements)) & sum(grepl("cpd:C00064_.",elements)))
{
for(i in 1:length(sub_net_transaminase[,1]))
{
for(j in 1:2)
{
if(grepl("C00025",sub_net_transaminase[i,j]) | grepl("C00064",sub_net_transaminase[i,j]))
{
sub_net_transaminase[i,columns[-j]] <- paste(sub_net_transaminase[i,columns[-j]],"_glugln",sep = "")
}
}
}
}
return(sub_net_transaminase)
}, transaminases_net = transaminases_net, USE.NAMES = T, simplify = F)
splitted_transaminase <- as.data.frame(do.call(rbind,splitted_transaminase))
sub_network_nocofact$reaction_network <- as.data.frame(rbind(sub_net_no_transaminase, splitted_transaminase))
sub_network_nocofact$reaction_network <- unique(sub_network_nocofact$reaction_network)
network_attributes <- as.data.frame(cbind(unique(c(sub_network_nocofact$reaction_network[,1],sub_network_nocofact$reaction_network[,2])),NA))
network_attributes[,2] <- ifelse(grepl("cpd:",network_attributes[,1]), "metabolite","reaction_enzyme")
names(network_attributes) <- names(sub_network_nocofact$attributes)
sub_network_nocofact$attributes <- network_attributes
return(sub_network_nocofact)
}
#'\code{nitrogen_tracking}
#'
#' This function cross link transamination metabolites to track amine groups
#'
#' @param sub_network_nocofact ipsum...
#' @return ipsum...
#' @export
nitrogen_tracking <- function(sub_network_nocofact)
{
sub_net <- sub_network_nocofact$reaction_network
transaminases_net <- sub_net[grepl("C0002[56]",sub_net$source) |
grepl("C0002[56]",sub_net$target),]
transaminases_potential <- unique(c(transaminases_net$source,transaminases_net$target))
transaminases_potential <- transaminases_potential[!grepl("cpd:",transaminases_potential)]
transaminases_net <- sub_net[sub_net$source %in% transaminases_potential |
sub_net$target %in% transaminases_potential,]
sub_net_no_transaminase <- sub_net[!(sub_net$source %in% transaminases_potential |
sub_net$target %in% transaminases_potential),]
splitted_transaminase <- sapply(transaminases_potential, function(transaminase, transaminases_net){
sub_net_transaminase <- transaminases_net[transaminase == transaminases_net$source |
transaminase == transaminases_net$target,]
columns <- c(1,2)
elements <- unique(c(sub_net_transaminase[,1],sub_net_transaminase[,2]))
if(sum(grepl("cpd:C00025_.",elements)) & sum(grepl("cpd:C00026_.",elements)))
{
if(sum(grepl("C00025",sub_net_transaminase[,1])) > 0)
{
for(i in 1:length(sub_net_transaminase[,1]))
{
if(grepl("C00025",sub_net_transaminase[i,1]))
{
sub_net_transaminase[i,2] <- paste(sub_net_transaminase[i,2],"_nitro",sep = "")
}
if(!grepl("C00026",sub_net_transaminase[i,2]) & grepl("cpd:C",sub_net_transaminase[i,2]))
{
sub_net_transaminase[i,1] <- paste(sub_net_transaminase[i,1],"_nitro",sep = "")
}
}
}else
{
for(i in 1:length(sub_net_transaminase[,1]))
{
if(grepl("C00025",sub_net_transaminase[i,2]))
{
sub_net_transaminase[i,1] <- paste(sub_net_transaminase[i,1],"_nitro",sep = "")
}
if(!grepl("C00026",sub_net_transaminase[i,1]) & grepl("cpd:C",sub_net_transaminase[i,1]))
{
sub_net_transaminase[i,2] <- paste(sub_net_transaminase[i,2],"_nitro",sep = "")
}
}
}
}
return(sub_net_transaminase)
}, transaminases_net = transaminases_net, USE.NAMES = T, simplify = F)
splitted_transaminase <- as.data.frame(do.call(rbind,splitted_transaminase))
sub_network_nocofact$reaction_network <- as.data.frame(rbind(sub_net_no_transaminase, splitted_transaminase))
sub_network_nocofact$reaction_network <- unique(sub_network_nocofact$reaction_network)
network_attributes <- as.data.frame(cbind(unique(c(sub_network_nocofact$reaction_network[,1],sub_network_nocofact$reaction_network[,2])),NA))
network_attributes[,2] <- ifelse(grepl("cpd:",network_attributes[,1]), "metabolite","reaction_enzyme")
names(network_attributes) <- names(sub_network_nocofact$attributes)
sub_network_nocofact$attributes <- network_attributes
return(sub_network_nocofact)
}
saezlab/ocean documentation built on Nov. 12, 2024, 5 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions nitrogen_tracking split_transaminases remove_cofactors model_to_pathway_sif
Documented in model_to_pathway_sif nitrogen_tracking remove_cofactors split_transaminases
#'\code{model_to_pathway_sif}
#'
#' This function generate a subreaction network in sif format from the
#' recon_redhuman model for a given set of pathways
#'
#' @param pathway_to_keep character vector with the ID of the pathways that
#' should be used to build the SIF reaction network.
#' Use "View(unique(recon2_redhuman$pathway))" to see available pathway IDs
#' @param pathways pathway corresponding to redhuman reaction indexes
#' @param rxns reaction IDs corresponding to redhuman reaction indexes
#' @param rules gene rules corresponding to redhuman reaction indexes
#' @param minmax min max flux boundaries corresponding to redhuman reaction indexes
#' @param stochio redhuman stochiometric matrix
#' @param metab metabolite IDs corresponding to redhuman metabolite indexes
#' @param mapping obsolete
#' @param recon1_metabolites obsolete
#' @return a subreaction network in sif format
#' @export
model_to_pathway_sif <- function(pathway_to_keep,
pathways = recon2_redhuman$pathway, #redHuman_models folder
rxns = recon2_redhuman$rxns,
rules = recon2_redhuman$rules,
minmax = recon2_redhuman$minmax,
stochio = recon2_redhuman$stochio,
metab = recon2_redhuman$metab,
mapping = recon2_redhuman$gene_mapping,
recon1_metabolites = recon2_redhuman$metab_mapping){
####
index_to_keep <- sapply(pathway_to_keep, function(x, pathways){
which(pathways == x)
}, pathways = pathways, simplify = T)
index_to_keep <- unique(unlist(index_to_keep))
####
rxns <- rxns[index_to_keep,, drop = F]
rules <- rules[index_to_keep,, drop = F]
minmax <- minmax[index_to_keep,, drop = F]
pathways <- pathways[index_to_keep,, drop = F]
stochio <- stochio[,index_to_keep,drop = F]
####
print("Building gprs...")
gprs <- build_grps(rxns, rules, minmax, pathways)
####
print("Building reaction list...")
reactions_list <- build_reactions_list(gprs, stochio, metab, minmax)
###
print("Building reaction data frame...")
reactions_df <- build_reactions_df(reactions_list, mapping)
###
# print("Converting network gene IDs...")
# reactions_df <- translate_network_gene_ids(reactions_df, mapping)
reactions_df <- reactions_df[!grepl("HC0",reactions_df$V1) & !grepl("HC0",reactions_df$V2),]
reactions_df <- reactions_df[complete.cases(reactions_df),]
###
print("Converting network metabolite IDs...")
reactions_df <- translate_network_metab_ids(reactions_df, recon1_metabolites)
names(reactions_df) <- c("source","target")
reactions_df <- unique(reactions_df)
###
row.names(reactions_df) <- c(1:length(reactions_df[,1]))
if("Urea cycle" %in% pathway_to_keep)
{
print("Ataching Urea cycle additional reactions")
reactions_df <- as.data.frame(rbind(reactions_df,network_supplements[["Urea cycle"]]))
}
if("Purine synthesis" %in% pathway_to_keep)
{
print("Ataching Purine synthesis additional reactions")
reactions_df <- as.data.frame(rbind(reactions_df,network_supplements[["Purine synthesis"]]))
}
if("Valine, leucine, and isoleucine metabolism" %in% pathway_to_keep)
{
print("Ataching BCAA additional reactions")
reactions_df <- as.data.frame(rbind(reactions_df,network_supplements[["Carnitine synthesis"]])) #add ketoacid to carnitine
reactions_df <- as.data.frame(rbind(reactions_df,network_supplements[["Valine leucine and isoleucine metabolism"]])) #add bcat1 reverse
# reactions_df <- reactions_df[reactions_df[,1] != "1629_594_593_1738" & #remove the wrong BCKDH complex and correct it
# reactions_df[,2] != "1629_594_593_1738" &
# reactions_df[,1] != "1629_1738_594_593" &
# reactions_df[,2] != "1629_1738_594_593",]
reactions_df <- reactions_df[!grepl("CRAT.*_reverse",reactions_df[,1]) & #remove reverse crat
!grepl("CRAT.*_reverse",reactions_df[,2]),]
# reactions_df <- as.data.frame(rbind(reactions_df,network_supplements[["BCKDH"]])) #add correct BCKDH
}
###
return(reactions_df)
}
#'\code{remove_cofactors}
#'
#' This function removes cofactors from reaction networks generate by the model_to_pathway_sif function
#'
#' @param reaction_network reaction network in SIF format,
#' like the one returned by model_to_pathway_sif
#' @param compound_list KEGGREST coumpound list
#' @return reaction network in SIF format and the corresponding
#' attribute dataframe
#' @export
remove_cofactors <- function(reaction_network, compound_list = kegg_compounds)
{
compounds <- unique(c(reaction_network$source, reaction_network$target))
compounds <- compounds[grepl("cpd[:]",compounds)]
compounds <- as.data.frame(compounds)
names(compounds)[1] <- "compounds_compartment"
compounds$compound <- gsub("_.*","",compounds$compounds_compartment)
compound_vec <- unique(compounds$compound)
bad_kegg_compounds <- list()
for (compound in compound_list)
{
brite <- compound$BRITE
compound$NAME <- gsub(";","",compound$NAME)
if(!is.null(compound$ATOM))
{
brite <- gsub("[ ]+","",brite)
if(("Cofactors" %in% brite |
"Nucleotides" %in% brite |
"CO2" %in% compound$NAME |
as.numeric(compound$ATOM[1]) <= 3 |
"ITP" %in% compound$NAME |
"IDP" %in% compound$NAME |
"NADH" %in% compound$NAME |
"NADPH" %in% compound$NAME |
"FADH2" %in% compound$NAME) &
!"SAM" %in% compound$NAME)
{
bad_kegg_compounds[[compound$ENTRY[1]]] <- compound
}
}
}
bad_entries <- c()
for(compound in bad_kegg_compounds)
{
bad_entries <- c(bad_entries,compound$ENTRY[1])
}
bad_entries <- paste("cpd:",bad_entries,sep = "")
bad_compounds <- compounds[compounds$compound %in% bad_entries,]
##manually add diphosphate to be removed from network
bad_compounds <- as.data.frame(rbind(bad_compounds,c("cpd:C00013_c","cpd:C00013")))
bad_compounds <- as.data.frame(rbind(bad_compounds,c("cpd:C00013_m","cpd:C00013")))
bad_compounds <- as.data.frame(rbind(bad_compounds,c("cpd:C00009_c","cpd:C00009")))
bad_compounds <- as.data.frame(rbind(bad_compounds,c("cpd:C00009_m","cpd:C00009")))
bad_compounds <- as.data.frame(rbind(bad_compounds,c("cpd:C00009_r","cpd:C00009")))
reaction_network_no_cofact <- reaction_network[
!(reaction_network$source %in% bad_compounds$compounds_compartment) &
!(reaction_network$target %in% bad_compounds$compounds_compartment),
]
reaction_network_no_cofact <-
reaction_network_no_cofact[
complete.cases(reaction_network_no_cofact),
]
nodes <- unique(c(reaction_network_no_cofact$source ,
reaction_network_no_cofact$target ))
node_attributes <- as.data.frame(matrix(NA,length(nodes),2))
node_attributes[,1] <- nodes
node_attributes[,2] <- "reaction_enzyme"
node_attributes[node_attributes[,1] %in% compounds$compounds_compartment,2] <- "metabolite"
return(list("reaction_network" = reaction_network_no_cofact, "attributes" = node_attributes))
}
#'\code{compress_transporters}
#'
#' This function removes cofactors from reaction networks generate by the model_to_pathway_sif function
#'
#' @param sub_network_nocofact ipsum...
#' @return ipsum...
#' @export
compress_transporters <- function (sub_network_nocofact)
{
test_1 <- paste(gsub("_[cxrnme]$", "", sub_network_nocofact$reaction_network$source),
gsub("_[cxrnme]$", "", sub_network_nocofact$reaction_network$target),
sep = "_")
test_2 <- paste(gsub("_[cxrnme]$", "", sub_network_nocofact$reaction_network$target),
gsub("_[cxrnme]$", "", sub_network_nocofact$reaction_network$source),
sep = "_")
test_1[test_1 %in% test_2]
transporters <- unique(c(sub_network_nocofact$reaction_network[test_1 %in%
test_2, 1], sub_network_nocofact$reaction_network[test_1 %in%
test_2, 2]))
transporters <- transporters[!grepl("_[cxrnme]$", transporters)]
for (i in 1:2) {
sub_network_nocofact$reaction_network[, i] <- sapply(sub_network_nocofact$reaction_network[,
i], function(x, transporters) {
if (x %in% transporters) {
return("transporter")
}
else {
return(x)
}
}, transporters = transporters)
}
sub_network_nocofact$reaction_network$edgeId <- paste(sub_network_nocofact$reaction_network$source,
sub_network_nocofact$reaction_network$target, sep = "_")
sub_network_nocofact$reaction_network <- sub_network_nocofact$reaction_network[!duplicated(sub_network_nocofact$reaction_network$edgeId),
]
edge_transporter <- sub_network_nocofact$reaction_network$edgeId
edge_transporter <- edge_transporter[grepl("transporter",
edge_transporter)]
groups <- rep(0, length(edge_transporter))
metab_memory <- c()
for (i in 1:length(edge_transporter)) {
if (grepl("^cpd:", edge_transporter[i]) | grepl("_[cxrnme]_",
edge_transporter[i])) {
metab <- gsub("_.*", "", edge_transporter[i])
if(!metab %in% metab_memory)
{
metab_memory[i] <- metab
for (j in 1:length(edge_transporter)) {
if (grepl(metab, edge_transporter[j]))
{
groups[j] <- i
}
}
}
}
}
names(groups) <- edge_transporter
network <- sub_network_nocofact$reaction_network
for (i in 1:2) {
for (j in 1:length(network[, i])) {
if (network[j, i] == "transporter") {
print(network[j, 3])
network[j, i] <- paste(network[j, i], groups[network[j,
3]], sep = ">")
}
}
}
network <- network[!grepl("^SLC", network$edgeId) & !grepl("_SLC",
network$edgeId), ]
sub_network_nocofact$reaction_network <- network[, -3]
sub_network_nocofact$attributes <- sub_network_nocofact$attributes[sub_network_nocofact$attributes[, 1] %in% network[, 1] |
sub_network_nocofact$attributes[, 1] %in% network[, 2], ]
new_transporters <- unique(c(network[, 1], network[, 2]))
new_transporters <- new_transporters[grepl("transporter",
new_transporters)]
new_transporters <- as.data.frame(cbind(new_transporters,
rep("transporter", length(new_transporters))))
names(new_transporters) <- c("V1", "V2")
sub_network_nocofact$attributes <- as.data.frame(rbind(sub_network_nocofact$attributes,
new_transporters))
return(sub_network_nocofact)
}
#'\code{split_transaminases}
#'
#' This function removes cofactors from reaction networks generate by the model_to_pathway_sif function
#'
#' @param sub_network_nocofact ipsum...
#' @return ipsum...
#' @export
split_transaminases <- function(sub_network_nocofact)
{
sub_net <- sub_network_nocofact$reaction_network
transaminases_net <- sub_net[grepl("C0002[56]",sub_net$source) |
grepl("C0002[56]",sub_net$target) |
grepl("C00064",sub_net$source) |
grepl("C00064",sub_net$target),]
sub_net_no_transaminase <- sub_net[!(grepl("C0002[56]",sub_net$source) |
grepl("C0002[56]",sub_net$target) |
grepl("C00064",sub_net$source) |
grepl("C00064",sub_net$target)),]
transaminases_potential <- unique(c(transaminases_net$source,transaminases_net$target))
transaminases_potential <- transaminases_potential[!grepl("cpd:",transaminases_potential)]
transaminases_net <- sub_net[sub_net$source %in% transaminases_potential |
sub_net$target %in% transaminases_potential,]
splitted_transaminase <- sapply(transaminases_potential, function(transaminase, transaminases_net){
sub_net_transaminase <- transaminases_net[transaminase == transaminases_net$source |
transaminase == transaminases_net$target,]
columns <- c(1,2)
elements <- unique(c(sub_net_transaminase[,1],sub_net_transaminase[,2]))
if(sum(grepl("cpd:C00025_.",elements)) & sum(grepl("cpd:C00026_.",elements)))
{
for(i in 1:length(sub_net_transaminase[,1]))
{
for(j in 1:2)
{
if(grepl("C0002[56]",sub_net_transaminase[i,j]))
{
sub_net_transaminase[i,columns[-j]] <- paste(sub_net_transaminase[i,columns[-j]],"_gluakg",sep = "")
}
}
}
}
if(sum(grepl("cpd:C00025_.",elements)) & sum(grepl("cpd:C00064_.",elements)))
{
for(i in 1:length(sub_net_transaminase[,1]))
{
for(j in 1:2)
{
if(grepl("C00025",sub_net_transaminase[i,j]) | grepl("C00064",sub_net_transaminase[i,j]))
{
sub_net_transaminase[i,columns[-j]] <- paste(sub_net_transaminase[i,columns[-j]],"_glugln",sep = "")
}
}
}
}
return(sub_net_transaminase)
}, transaminases_net = transaminases_net, USE.NAMES = T, simplify = F)
splitted_transaminase <- as.data.frame(do.call(rbind,splitted_transaminase))
sub_network_nocofact$reaction_network <- as.data.frame(rbind(sub_net_no_transaminase, splitted_transaminase))
sub_network_nocofact$reaction_network <- unique(sub_network_nocofact$reaction_network)
network_attributes <- as.data.frame(cbind(unique(c(sub_network_nocofact$reaction_network[,1],sub_network_nocofact$reaction_network[,2])),NA))
network_attributes[,2] <- ifelse(grepl("cpd:",network_attributes[,1]), "metabolite","reaction_enzyme")
names(network_attributes) <- names(sub_network_nocofact$attributes)
sub_network_nocofact$attributes <- network_attributes
return(sub_network_nocofact)
}
#'\code{nitrogen_tracking}
#'
#' This function cross link transamination metabolites to track amine groups
#'
#' @param sub_network_nocofact ipsum...
#' @return ipsum...
#' @export
nitrogen_tracking <- function(sub_network_nocofact)
{
sub_net <- sub_network_nocofact$reaction_network
transaminases_net <- sub_net[grepl("C0002[56]",sub_net$source) |
grepl("C0002[56]",sub_net$target),]
transaminases_potential <- unique(c(transaminases_net$source,transaminases_net$target))
transaminases_potential <- transaminases_potential[!grepl("cpd:",transaminases_potential)]
transaminases_net <- sub_net[sub_net$source %in% transaminases_potential |
sub_net$target %in% transaminases_potential,]
sub_net_no_transaminase <- sub_net[!(sub_net$source %in% transaminases_potential |
sub_net$target %in% transaminases_potential),]
splitted_transaminase <- sapply(transaminases_potential, function(transaminase, transaminases_net){
sub_net_transaminase <- transaminases_net[transaminase == transaminases_net$source |
transaminase == transaminases_net$target,]
columns <- c(1,2)
elements <- unique(c(sub_net_transaminase[,1],sub_net_transaminase[,2]))
if(sum(grepl("cpd:C00025_.",elements)) & sum(grepl("cpd:C00026_.",elements)))
{
if(sum(grepl("C00025",sub_net_transaminase[,1])) > 0)
{
for(i in 1:length(sub_net_transaminase[,1]))
{
if(grepl("C00025",sub_net_transaminase[i,1]))
{
sub_net_transaminase[i,2] <- paste(sub_net_transaminase[i,2],"_nitro",sep = "")
}
if(!grepl("C00026",sub_net_transaminase[i,2]) & grepl("cpd:C",sub_net_transaminase[i,2]))
{
sub_net_transaminase[i,1] <- paste(sub_net_transaminase[i,1],"_nitro",sep = "")
}
}
}else
{
for(i in 1:length(sub_net_transaminase[,1]))
{
if(grepl("C00025",sub_net_transaminase[i,2]))
{
sub_net_transaminase[i,1] <- paste(sub_net_transaminase[i,1],"_nitro",sep = "")
}
if(!grepl("C00026",sub_net_transaminase[i,1]) & grepl("cpd:C",sub_net_transaminase[i,1]))
{
sub_net_transaminase[i,2] <- paste(sub_net_transaminase[i,2],"_nitro",sep = "")
}
}
}
}
return(sub_net_transaminase)
}, transaminases_net = transaminases_net, USE.NAMES = T, simplify = F)
splitted_transaminase <- as.data.frame(do.call(rbind,splitted_transaminase))
sub_network_nocofact$reaction_network <- as.data.frame(rbind(sub_net_no_transaminase, splitted_transaminase))
sub_network_nocofact$reaction_network <- unique(sub_network_nocofact$reaction_network)
network_attributes <- as.data.frame(cbind(unique(c(sub_network_nocofact$reaction_network[,1],sub_network_nocofact$reaction_network[,2])),NA))
network_attributes[,2] <- ifelse(grepl("cpd:",network_attributes[,1]), "metabolite","reaction_enzyme")
names(network_attributes) <- names(sub_network_nocofact$attributes)
sub_network_nocofact$attributes <- network_attributes
return(sub_network_nocofact)
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.