R/renameTaxa.R
In stefpeschel/NetCoMi: Network Construction and Comparison for Microbiome Data
Defines functions
renameTaxa
Documented in
renameTaxa
#' @title Rename taxa
#'
#' @description Function for renaming taxa in a taxonomic table, which can be
#' given as matrix or phyloseq object. \cr\cr
#' It comes with functionality for making unknown
#' and unclassified taxa unique and substituting them by the next higher known
#' taxonomic level, e.g., an unknown genus "g__" can automatically be renamed
#' to "1_Streptococcaceae(F)".
#' User-defined patterns determine the format of known and substituted names.
#' Unknown names (e.g., NAs) and unclassified taxa can be
#' handled separately. Duplicated names within one or more chosen ranks can
#' also be made unique by numbering them consecutively.
#'
#' @param taxtab taxonomic table (matrix containing the taxonomic names; columns
#' must be taxonomic ranks) or phyloseq object.
#' @param pat character specifying the pattern of new taxonomic names if the
#' current name is KNOWN.
#' See the examples and default value for a demo.
#' Possible space holders are:
#' \describe{
#' \item{\code{<name>}}{Taxonomic name (either the original or replaced one)}
#' \item{\code{<rank>}}{Taxonomic rank in lower case}
#' \item{\code{<Rank>}}{Taxonomic rank with first letter in upper case}
#' \item{\code{<r>}}{Abbreviated taxonomic rank in lower case}
#' \item{\code{<R>}}{Abbreviated taxonomic rank in upper case}
#' }
#' @param substPat character specifying the pattern of new taxonomic names if the
#' current name is UNKNOWN. The current name is substituted by the next higher
#' existing name.
#' Possible space holders (in addition to that of \code{pat}):
#' \describe{
#' \item{\code{<subst_name>}}{Substituted taxonomic name (next higher existing
#' name)}
#' \item{\code{<subst_rank>}}{Taxonomic rank of substitute name in lower case}
#' \item{\code{<subst_Rank>}}{Taxonomic rank of substitute name with first letter
#' in upper case}
#' \item{\code{<subst_r>}}{Abbreviated taxonomic rank of substitute name in
#' lower case}
#' \item{\code{<subst_R>}}{Abbreviated taxonomic rank of substitute name in
#' upper case}
#' }
#' @param unknown character vector giving the labels of unknown taxa, without
#' leading rank label (e.g., "g_" or "g__" for genus level). If
#' \code{numUnknown = TRUE}, unknown names are replaced by a number.
#' @param numUnknown logical. If \code{TRUE}, a number is assigned to all
#' unknown taxonomic names (defined by \code{unknown}) to make them unique.
#' @param unclass character vector giving the label of unclassified taxa,
#' without leading rank label (e.g., "g_" or "g__" for genus level). If
#' \code{numUnclass = TRUE}, a number is added to the names of unclassified
#' taxa. Note that unclassified taxa and unknown taxa get a separate numbering
#' if \code{unclass} is set. To replace all unknown and unclassified taxa by
#' numbers, add "unclassified" (or the appropriate counterpart) to
#' \code{unknown} and set \code{unclass} to \code{NULL}.
#' @param numUnclass logical. If \code{TRUE}, a number is assigned to all
#' unclassified taxa (defined by \code{unclass}) to make them unique.
#' The pattern is defined via \code{numUnclassPat}.
#' @param numUnclassPat character defining the pattern used for numbering
#' unclassified taxa. Must include a space holder for the name ("<name>") and
#' one for the number ("<num>"). Default is "<name><num>" resulting e.g., in
#' "unclassified1".
#' @param numDupli character vector giving the ranks that should be made unique
#' by adding a number. Elements must match column names. The pattern is
#' defined via \code{numDupliPat}.
#' @param numDupliPat character defining the pattern used for numbering
#' duplicated names (if \code{numDupli} is given). Must include a space holder
#' for the name ("<name>") and one for the number ("<num>").
#' Default is "<name><num>" resulting e.g., in "Ruminococcus1".
#' @param ranks character vector giving rank names used for renaming the
#' taxa. If \code{NULL}, the functions tries to automatically set rank names
#' based on common usage.
#' @param ranksAbb character vector giving abbreviated rank names, which are
#' directly used for the place holders <r>, <subst_r>, <R>, and <subst_R>
#' (the former two in lower case and the latter two in upper case).
#' If \code{NULL}, the first letter of the rank names is used.
#' @param ignoreCols numeric vector with columns to be ignored. Names remain
#' unchanged for these columns. Columns containing \code{NA}s are ignored
#' automatically only if \code{ignoreCols = NULL}.
#' Note: length of \code{ranks} and \code{ranksAbb} must
#' match the number of non-ignored columns.
#' @return Renamed taxonomic table (matrix or phyloseq object, depending on the
#' input).
#'
#' @examples
#' #--- Load and edit data -----------------------------------------------------
#'
#' library(phyloseq)
#' data("GlobalPatterns")
#' global <- subset_taxa(GlobalPatterns, Kingdom == "Bacteria")
#' taxtab <- global@tax_table@.Data[1:10, ]
#'
#' # Add some unclassified taxa
#' taxtab[c(2,3,5), "Species"] <- "unclassified"
#' taxtab[c(2,3), "Genus"] <- "unclassified"
#' taxtab[2, "Family"] <- "unclassified"
#'
#' # Add some blanks
#' taxtab[7, "Genus"] <- " "
#' taxtab[7:9, "Species"] <- " "
#'
#' # Add taxon that is unclassified up to Kingdom
#' taxtab[9, ] <- "unclassified"
#' taxtab[9, 1] <- "Unclassified"
#'
#' # Add row names
#' rownames(taxtab) <- paste0("OTU", 1:nrow(taxtab))
#'
#' print(taxtab)
#'
#' #--- Example 1 (default setting) --------------------------------------------
#'
#' # Example 1 (default setting)
#' # - Known names are replaced by "<r>_<name>"
#' # - Unknown names are replaced by "<r>_<name>_<subst_r>_<subst_name>"
#' # - Unclassified taxa have separate numbering
#' # - Ranks are taken from column names
#' # - e.g., unknown genus -> "g_1_f_Streptococcaceae"
#'
#' renamed1 <- renameTaxa(taxtab)
#' renamed1
#'
#' #--- Example 2 --------------------------------------------------------------
#' # - Use phyloseq object (subset of class clostridia to decrease runtime)
#'
#' global_sub <- subset_taxa(global, Class == "Clostridia")
#'
#' renamed2 <- renameTaxa(global_sub)
#' tax_table(renamed2)[1:5, ]
#'
#' #--- Example 3 --------------------------------------------------------------
#' # - Known names remain unchanged
#' # - Substituted names are indicated by their rank in brackets
#' # - Pattern for numbering unclassified taxa changed
#' # - e.g., unknown genus -> "Streptococcaceae (F)"
#' # - Note: Numbering of unknowns is not shown because "<name>" is not
#' # included in "substPat"
#'
#' renamed3 <- renameTaxa(taxtab, numUnclassPat = "<name>_<num>",
#' pat = "<name>",
#' substPat = "<subst_name> (<subst_R>)")
#' renamed3
#'
#' #--- Example 4 --------------------------------------------------------------
#' # - Same as before but numbering shown for unknown names
#' # - e.g., unknown genus -> "1 Streptococcaceae (F)"
#'
#' renamed4 <- renameTaxa(taxtab, numUnclassPat = "<name>_<num>",
#' pat = "<name>",
#' substPat = "<name> <subst_name> (<subst_R>)")
#' renamed4
#'
#' #--- Example 5 --------------------------------------------------------------
#' # - Same numbering for unkown names and unclassified taxa
#' # - e.g., unknown genus -> "1_Streptococcaceae(F)"
#' # - Note: We get a warning here because "Unclassified" (with capital U)
#' # are not included in "unknown" but occur in the data
#'
#' renamed5 <- renameTaxa(taxtab, unclass = NULL,
#' unknown = c(NA, " ", "unclassified"),
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_R>)")
#' renamed5
#'
#' #--- Example 6 --------------------------------------------------------------
#' # - Same as before, but OTU9 is now renamed correctly
#'
#' renamed6 <- renameTaxa(taxtab, unclass = NULL,
#' unknown = c(NA, " ", "unclassified", "Unclassified"),
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_R>)")
#' renamed6
#'
#' #--- Example 7 --------------------------------------------------------------
#' # - Add "(<Rank>: unknown)" to unknown names
#' # - e.g., unknown genus -> "1 Streptococcaceae (Genus: unknown)"
#'
#' renamed7 <- renameTaxa(taxtab, unclass = NULL,
#' unknown = c(NA, " ", "unclassified", "Unclassified"),
#' pat = "<name>",
#' substPat = "<name> <subst_name> (<Rank>: unknown)")
#' renamed7
#'
#' #--- Example 8 --------------------------------------------------------------
#' # - Do not substitute unknowns and unclassified taxa by higher ranks
#' # - e.g., unknown genus -> "1"
#'
#' renamed8 <- renameTaxa(taxtab,
#' pat = "<name>", substPat = "<name>")
#' renamed8
#'
#' #--- Example 9 --------------------------------------------------------------
#' # - Error if ranks cannot be automatically determined
#' # from column names or taxonomic names
#'
#' taxtab_noranks <- taxtab
#' colnames(taxtab_noranks) <- paste0("Rank", 1:ncol(taxtab))
#' head(taxtab_noranks)
#'
#' \dontrun{
#' renamed9 <- renameTaxa(taxtab_noranks,
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_R>)")
#' }
#'
#' # Ranks can either be given via "ranks" ...
#' (ranks <- colnames(taxtab))
#'
#' renamed9 <- renameTaxa(taxtab_noranks,
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_R>)",
#' ranks = ranks)
#' renamed9
#'
#' # ... or "ranksAbb" (we now use the lower case within "substPat")
#' (ranks <- substr(colnames(taxtab), 1, 1))
#'
#' renamed9 <- renameTaxa(taxtab_noranks,
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_r>)",
#' ranksAbb = ranks)
#' renamed9
#'
#' #--- Example 10 -------------------------------------------------------------
#' # - Make names of ranks "Family" and "Order" unique by adding numbers to
#' # duplicated names
#'
#' renamed10 <- renameTaxa(taxtab,
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_R>)",
#' numDupli = c("Family", "Order"))
#' renamed10
#'
#' any(duplicated(renamed10[, "Family"]))
#' any(duplicated(renamed10[, "Order"]))
#'
#' @importFrom phyloseq tax_table
#' @export
renameTaxa <- function(taxtab,
pat = "<r>_<name>",
substPat = "<r>_<name>_<subst_r>_<subst_name>",
unknown = c(NA, "", " ", "__"),
numUnknown = TRUE,
unclass = c("unclassified", "Unclassified"),
numUnclass = TRUE,
numUnclassPat = "<name><num>",
numDupli = NULL,
numDupliPat = "<name><num>",
ranks = NULL, ranksAbb = NULL,
ignoreCols = NULL) {
stopifnot(is.character(pat))
stopifnot(is.character(substPat))
stopifnot(is.logical(numUnknown))
stopifnot(is.logical(numUnclass))
if (inherits(taxtab, "phyloseq")) {
tax <- as.matrix(taxtab@tax_table@.Data)
} else {
tax <- as.matrix(taxtab)
}
# Input check for numUnclass and numDupli
if (numUnclass) {
if (!grepl("<name>", numUnclassPat)) {
stop('Argument "numUnclassPat" must contain a space holder "<name>".')
}
if (!grepl("<num>", numUnclassPat)) {
stop('Argument "numUnclassPat" must contain a space holder "<num>".')
}
}
if (!is.null(numDupli)) {
if (!all(numDupli %in% colnames(tax))) {
stop('Ranks given with "numDupli" must match column names of ' ,
'taxonomic table.')
}
if (!grepl("<name>", numDupliPat)) {
stop('Argument "numDupliPat" must contain a space holder "<name>".')
}
if (!grepl("<num>", numDupliPat)) {
stop('Argument "numDupliPat" must contain a space holder "<num>".')
}
}
# Remove columns to ignore
if (!is.null(ignoreCols)) {
if (is.character(ignoreCols)) {
ignoreCols <- which(colnames(tax) %in% ignoreCols)
}
} else {
# Ignore columns that contain NAs only
ignore.tmp <- which(apply(tax, 2, function(x) all(is.na(x))))
if (length(ignore.tmp) == 1) {
message(paste0("Column ", ignore.tmp,
" contains NAs only and is ignored."))
ignoreCols <- ignore.tmp
} else if (length(ignore.tmp) > 1) {
message(paste0("Columns ", ignore.tmp,
" contain NAs only and are ignored."))
ignoreCols <- ignore.tmp
}
}
tax_orig <- tax
if (!is.null(ignoreCols)) {
tax <- tax_orig[, -ignoreCols]
}
# Define ranks
nranks <- ncol(tax)
ranks.tmp <- c("Life", "Domain", "Kingdom", "Phylum", "Class", "Order",
"Family", "Genus", "Species")
ranksAbb.tmp <- substr(tolower(ranks.tmp), 1, 1)
missRanks <- missRanksAbb <- FALSE
if (is.null(ranks)) {
if (any(ranks.tmp %in% colnames(tax))) {
ranks <- colnames(tax)
} else {
repl <- 0
ranks <- rep(NA, ncol(tax))
for (r in seq_along(ranks.tmp)) {
sel <- grep(paste0(ranksAbb.tmp, "_")[r], t(tax)) %% nranks
if (length(sel) != 0) {
sel <- sel[1]
if (sel == 0) sel <- nranks
ranks[sel] <- ranks.tmp[r]
repl <- repl + 1
}
}
if (repl != nranks) {
missRanks <- TRUE
}
}
} else {
if (length(ranks) != nranks) {
stop("Length of 'ranks' does not match number of columns")
}
}
if (is.null(ranksAbb)) {
if (missRanks) {
missRanksAbb <- TRUE
} else {
ranksAbb <- substr(tolower(ranks), 1, 1)
}
} else {
if (length(ranksAbb) != nranks) {
stop("Length of 'ranksAbb' does not match number of columns")
}
}
# Stop if full rank names are not given but needed
if (missRanks && (grepl("<rank>", pat) |
grepl("<Rank>", pat))) {
stop("Ranks could not be determined but are needed for the given pattern. ",
"Please provide 'ranks'")
}
if (missRanks && (grepl("<rank>", substPat) |
grepl("<Rank>", substPat) |
grepl("<subst_rank>", substPat) |
grepl("<subst_Rank>", substPat))) {
stop("Ranks could not be determined but are needed for the given ",
"substitute pattern. Please provide 'ranks'")
}
# Stop if abbreviated rank names are not given but needed
if (missRanksAbb && (grepl("<r>", pat) |
grepl("<R>", pat))) {
stop("Abbreviated ranks could not be determined but are needed for the ",
"given pattern. Please provide 'ranks' or 'ranksAbb'.")
}
if (missRanksAbb && (grepl("<r>", substPat) |
grepl("<R>", substPat) |
grepl("<subst_r>", substPat) |
grepl("<subst_R>", substPat))) {
stop("Ranks could not be determined but are needed for the given ",
"substitute pattern. Please provide 'ranks' or 'ranksAbb'.")
}
for (r in seq_along(ranksAbb.tmp)) {
tax <- gsub(paste0(ranksAbb.tmp[r], "__"), "", tax)
tax <- gsub(paste0(ranksAbb.tmp[r], "_"), "", tax)
}
#-----------------------------------------------------------------------------
### Make labels unique
unknownMat <- matrix(data = FALSE, nrow = nrow(tax), ncol = nranks)
dimnames(unknownMat) <- dimnames(tax)
# Add number to unclassified taxa
if (!is.null(unclass)) {
if (numUnclass) {
# Split pattern
pat.tmp <- unlist(strsplit(numUnclassPat, "<"))
pat.tmp <- unlist(strsplit(pat.tmp, ">"))
numUnclassPat <- pat.tmp
for (c in 1:nranks) {
# Detect unclassified taxa
isunclass <- tax[, c] %in% unclass
if (sum(isunclass) > 0) {
# Replace names according to pattern
tmp <- matrix(rep(numUnclassPat,
each = sum(isunclass)), nrow = sum(isunclass))
tmp[tmp == "name"] <- tax[isunclass, c]
tmp[tmp == "num"] <- 1:sum(isunclass)
tax[isunclass, c] <- apply(tmp, 1, paste, collapse="")
unknownMat[isunclass, c] <- TRUE
}
}
}
} else {
if (any(grepl("unclass", tax, ignore.case = TRUE))) {
unclassname <- unique(tax[grep("unclass", tax, ignore.case = TRUE)])
if (!all(unclassname %in% unknown)) {
unclassname <- unclassname[!unclassname %in% unknown]
warning(paste0('Taxonomic table contains unclassified taxa. ',
'Consider adding "', unclassname,
'" to argument "unknown".'))
}
}
}
# Replace unknowns by number
if (numUnknown) {
for (c in 1:nranks) {
# Add a number if the rank is unknown
isunknown <- tax[, c] %in% unknown
if (sum(isunknown) > 0) {
tax[isunknown, c] <- 1:sum(isunknown)
unknownMat[isunknown, c] <- TRUE
}
}
}
# Make duplicates unique
if (!is.null(numDupli)) {
# Split pattern
pat.tmp <- unlist(strsplit(numDupliPat, "<"))
pat.tmp <- unlist(strsplit(pat.tmp, ">"))
numDupliPat <- pat.tmp
for (c in seq_along(numDupli)) {
tax.tmp <- tax[, numDupli[c]]
dupli <- table(tax.tmp)[table(tax.tmp) > 1]
if (length(dupli) > 0) {
for (d in seq_along(dupli)) {
# Replace names according to pattern
tmp <- matrix(rep(numDupliPat, each = dupli[d]), nrow = dupli[d])
tmp[tmp == "name"] <- names(dupli[d])
tmp[tmp == "num"] <- 1:dupli[d]
tax.tmp[tax.tmp == names(dupli[d])] <- apply(tmp, 1, paste,
collapse="")
}
tax[, numDupli[c]] <- tax.tmp
}
}
}
#-----------------------------------------------------------------------------
pat.tmp <- unlist(strsplit(pat, "<"))
pat.tmp <- unlist(strsplit(pat.tmp, ">"))
pat <- pat.tmp
pat.tmp <- unlist(strsplit(substPat, "<"))
pat.tmp <- unlist(strsplit(pat.tmp, ">"))
substPat <- pat.tmp
for (i in 1:nrow(tax)) {
for (r in nranks:1) {
name <- tax[i, r]
isunknown <- unknownMat[i, r]
usesubst <- NA
if (isunknown) {
usesubst <- TRUE
# Which ranks are known?
knownr <- which(!unknownMat[i, ])
if (length(knownr) == 0) {
# If all ranks are unknown, take name of the highest rank
sr <- 1
if (r == 1) usesubst <- FALSE
} else {
# Select last known rank
sr <- knownr[length(knownr)]
}
} else {
usesubst <- FALSE
}
# Assign the right pattern
if (usesubst) {
patout <- substPat
} else {
patout <- pat
}
patout[patout == "r"] <- tolower(ranksAbb[r])
patout[patout == "R"] <- toupper(ranksAbb[r])
patout[patout == "rank"] <- tolower(ranks[r])
patout[patout == "Rank"] <- paste0(toupper(substr(ranks[r], 1, 1)),
substr(ranks[r], 2,
nchar(ranks[r])))
patout[patout == "name"] <- name
if (usesubst) {
patout[patout == "subst_r"] <- tolower(ranksAbb[sr])
patout[patout == "subst_R"] <- toupper(ranksAbb[sr])
patout[patout == "subst_rank"] <- tolower(ranks[sr])
patout[patout == "subst_Rank"] <- paste0(toupper(substr(ranks[sr], 1, 1)),
substr(ranks[sr], 2,
nchar(ranks[sr])))
patout[patout == "subst_name"] <- tax[i, sr]
}
taxname <- paste(patout, collapse = "")
tax[i, r] <- taxname
}
}
if (!is.null(ignoreCols)) {
taxout <- tax_orig
taxout[, (1:ncol(tax_orig))[-ignoreCols]] <- tax
} else {
taxout <- tax
}
if (inherits(taxtab, "phyloseq")) {
phyloseq::tax_table(taxtab) <- phyloseq::tax_table(taxout)
taxout <- taxtab
}
return(taxout)
}
stefpeschel/NetCoMi documentation built on Nov. 12, 2024, 7:12 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 renameTaxa
Documented in renameTaxa
#' @title Rename taxa
#'
#' @description Function for renaming taxa in a taxonomic table, which can be
#' given as matrix or phyloseq object. \cr\cr
#' It comes with functionality for making unknown
#' and unclassified taxa unique and substituting them by the next higher known
#' taxonomic level, e.g., an unknown genus "g__" can automatically be renamed
#' to "1_Streptococcaceae(F)".
#' User-defined patterns determine the format of known and substituted names.
#' Unknown names (e.g., NAs) and unclassified taxa can be
#' handled separately. Duplicated names within one or more chosen ranks can
#' also be made unique by numbering them consecutively.
#'
#' @param taxtab taxonomic table (matrix containing the taxonomic names; columns
#' must be taxonomic ranks) or phyloseq object.
#' @param pat character specifying the pattern of new taxonomic names if the
#' current name is KNOWN.
#' See the examples and default value for a demo.
#' Possible space holders are:
#' \describe{
#' \item{\code{<name>}}{Taxonomic name (either the original or replaced one)}
#' \item{\code{<rank>}}{Taxonomic rank in lower case}
#' \item{\code{<Rank>}}{Taxonomic rank with first letter in upper case}
#' \item{\code{<r>}}{Abbreviated taxonomic rank in lower case}
#' \item{\code{<R>}}{Abbreviated taxonomic rank in upper case}
#' }
#' @param substPat character specifying the pattern of new taxonomic names if the
#' current name is UNKNOWN. The current name is substituted by the next higher
#' existing name.
#' Possible space holders (in addition to that of \code{pat}):
#' \describe{
#' \item{\code{<subst_name>}}{Substituted taxonomic name (next higher existing
#' name)}
#' \item{\code{<subst_rank>}}{Taxonomic rank of substitute name in lower case}
#' \item{\code{<subst_Rank>}}{Taxonomic rank of substitute name with first letter
#' in upper case}
#' \item{\code{<subst_r>}}{Abbreviated taxonomic rank of substitute name in
#' lower case}
#' \item{\code{<subst_R>}}{Abbreviated taxonomic rank of substitute name in
#' upper case}
#' }
#' @param unknown character vector giving the labels of unknown taxa, without
#' leading rank label (e.g., "g_" or "g__" for genus level). If
#' \code{numUnknown = TRUE}, unknown names are replaced by a number.
#' @param numUnknown logical. If \code{TRUE}, a number is assigned to all
#' unknown taxonomic names (defined by \code{unknown}) to make them unique.
#' @param unclass character vector giving the label of unclassified taxa,
#' without leading rank label (e.g., "g_" or "g__" for genus level). If
#' \code{numUnclass = TRUE}, a number is added to the names of unclassified
#' taxa. Note that unclassified taxa and unknown taxa get a separate numbering
#' if \code{unclass} is set. To replace all unknown and unclassified taxa by
#' numbers, add "unclassified" (or the appropriate counterpart) to
#' \code{unknown} and set \code{unclass} to \code{NULL}.
#' @param numUnclass logical. If \code{TRUE}, a number is assigned to all
#' unclassified taxa (defined by \code{unclass}) to make them unique.
#' The pattern is defined via \code{numUnclassPat}.
#' @param numUnclassPat character defining the pattern used for numbering
#' unclassified taxa. Must include a space holder for the name ("<name>") and
#' one for the number ("<num>"). Default is "<name><num>" resulting e.g., in
#' "unclassified1".
#' @param numDupli character vector giving the ranks that should be made unique
#' by adding a number. Elements must match column names. The pattern is
#' defined via \code{numDupliPat}.
#' @param numDupliPat character defining the pattern used for numbering
#' duplicated names (if \code{numDupli} is given). Must include a space holder
#' for the name ("<name>") and one for the number ("<num>").
#' Default is "<name><num>" resulting e.g., in "Ruminococcus1".
#' @param ranks character vector giving rank names used for renaming the
#' taxa. If \code{NULL}, the functions tries to automatically set rank names
#' based on common usage.
#' @param ranksAbb character vector giving abbreviated rank names, which are
#' directly used for the place holders <r>, <subst_r>, <R>, and <subst_R>
#' (the former two in lower case and the latter two in upper case).
#' If \code{NULL}, the first letter of the rank names is used.
#' @param ignoreCols numeric vector with columns to be ignored. Names remain
#' unchanged for these columns. Columns containing \code{NA}s are ignored
#' automatically only if \code{ignoreCols = NULL}.
#' Note: length of \code{ranks} and \code{ranksAbb} must
#' match the number of non-ignored columns.
#' @return Renamed taxonomic table (matrix or phyloseq object, depending on the
#' input).
#'
#' @examples
#' #--- Load and edit data -----------------------------------------------------
#'
#' library(phyloseq)
#' data("GlobalPatterns")
#' global <- subset_taxa(GlobalPatterns, Kingdom == "Bacteria")
#' taxtab <- global@tax_table@.Data[1:10, ]
#'
#' # Add some unclassified taxa
#' taxtab[c(2,3,5), "Species"] <- "unclassified"
#' taxtab[c(2,3), "Genus"] <- "unclassified"
#' taxtab[2, "Family"] <- "unclassified"
#'
#' # Add some blanks
#' taxtab[7, "Genus"] <- " "
#' taxtab[7:9, "Species"] <- " "
#'
#' # Add taxon that is unclassified up to Kingdom
#' taxtab[9, ] <- "unclassified"
#' taxtab[9, 1] <- "Unclassified"
#'
#' # Add row names
#' rownames(taxtab) <- paste0("OTU", 1:nrow(taxtab))
#'
#' print(taxtab)
#'
#' #--- Example 1 (default setting) --------------------------------------------
#'
#' # Example 1 (default setting)
#' # - Known names are replaced by "<r>_<name>"
#' # - Unknown names are replaced by "<r>_<name>_<subst_r>_<subst_name>"
#' # - Unclassified taxa have separate numbering
#' # - Ranks are taken from column names
#' # - e.g., unknown genus -> "g_1_f_Streptococcaceae"
#'
#' renamed1 <- renameTaxa(taxtab)
#' renamed1
#'
#' #--- Example 2 --------------------------------------------------------------
#' # - Use phyloseq object (subset of class clostridia to decrease runtime)
#'
#' global_sub <- subset_taxa(global, Class == "Clostridia")
#'
#' renamed2 <- renameTaxa(global_sub)
#' tax_table(renamed2)[1:5, ]
#'
#' #--- Example 3 --------------------------------------------------------------
#' # - Known names remain unchanged
#' # - Substituted names are indicated by their rank in brackets
#' # - Pattern for numbering unclassified taxa changed
#' # - e.g., unknown genus -> "Streptococcaceae (F)"
#' # - Note: Numbering of unknowns is not shown because "<name>" is not
#' # included in "substPat"
#'
#' renamed3 <- renameTaxa(taxtab, numUnclassPat = "<name>_<num>",
#' pat = "<name>",
#' substPat = "<subst_name> (<subst_R>)")
#' renamed3
#'
#' #--- Example 4 --------------------------------------------------------------
#' # - Same as before but numbering shown for unknown names
#' # - e.g., unknown genus -> "1 Streptococcaceae (F)"
#'
#' renamed4 <- renameTaxa(taxtab, numUnclassPat = "<name>_<num>",
#' pat = "<name>",
#' substPat = "<name> <subst_name> (<subst_R>)")
#' renamed4
#'
#' #--- Example 5 --------------------------------------------------------------
#' # - Same numbering for unkown names and unclassified taxa
#' # - e.g., unknown genus -> "1_Streptococcaceae(F)"
#' # - Note: We get a warning here because "Unclassified" (with capital U)
#' # are not included in "unknown" but occur in the data
#'
#' renamed5 <- renameTaxa(taxtab, unclass = NULL,
#' unknown = c(NA, " ", "unclassified"),
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_R>)")
#' renamed5
#'
#' #--- Example 6 --------------------------------------------------------------
#' # - Same as before, but OTU9 is now renamed correctly
#'
#' renamed6 <- renameTaxa(taxtab, unclass = NULL,
#' unknown = c(NA, " ", "unclassified", "Unclassified"),
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_R>)")
#' renamed6
#'
#' #--- Example 7 --------------------------------------------------------------
#' # - Add "(<Rank>: unknown)" to unknown names
#' # - e.g., unknown genus -> "1 Streptococcaceae (Genus: unknown)"
#'
#' renamed7 <- renameTaxa(taxtab, unclass = NULL,
#' unknown = c(NA, " ", "unclassified", "Unclassified"),
#' pat = "<name>",
#' substPat = "<name> <subst_name> (<Rank>: unknown)")
#' renamed7
#'
#' #--- Example 8 --------------------------------------------------------------
#' # - Do not substitute unknowns and unclassified taxa by higher ranks
#' # - e.g., unknown genus -> "1"
#'
#' renamed8 <- renameTaxa(taxtab,
#' pat = "<name>", substPat = "<name>")
#' renamed8
#'
#' #--- Example 9 --------------------------------------------------------------
#' # - Error if ranks cannot be automatically determined
#' # from column names or taxonomic names
#'
#' taxtab_noranks <- taxtab
#' colnames(taxtab_noranks) <- paste0("Rank", 1:ncol(taxtab))
#' head(taxtab_noranks)
#'
#' \dontrun{
#' renamed9 <- renameTaxa(taxtab_noranks,
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_R>)")
#' }
#'
#' # Ranks can either be given via "ranks" ...
#' (ranks <- colnames(taxtab))
#'
#' renamed9 <- renameTaxa(taxtab_noranks,
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_R>)",
#' ranks = ranks)
#' renamed9
#'
#' # ... or "ranksAbb" (we now use the lower case within "substPat")
#' (ranks <- substr(colnames(taxtab), 1, 1))
#'
#' renamed9 <- renameTaxa(taxtab_noranks,
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_r>)",
#' ranksAbb = ranks)
#' renamed9
#'
#' #--- Example 10 -------------------------------------------------------------
#' # - Make names of ranks "Family" and "Order" unique by adding numbers to
#' # duplicated names
#'
#' renamed10 <- renameTaxa(taxtab,
#' pat = "<name>",
#' substPat = "<name>_<subst_name>(<subst_R>)",
#' numDupli = c("Family", "Order"))
#' renamed10
#'
#' any(duplicated(renamed10[, "Family"]))
#' any(duplicated(renamed10[, "Order"]))
#'
#' @importFrom phyloseq tax_table
#' @export
renameTaxa <- function(taxtab,
pat = "<r>_<name>",
substPat = "<r>_<name>_<subst_r>_<subst_name>",
unknown = c(NA, "", " ", "__"),
numUnknown = TRUE,
unclass = c("unclassified", "Unclassified"),
numUnclass = TRUE,
numUnclassPat = "<name><num>",
numDupli = NULL,
numDupliPat = "<name><num>",
ranks = NULL, ranksAbb = NULL,
ignoreCols = NULL) {
stopifnot(is.character(pat))
stopifnot(is.character(substPat))
stopifnot(is.logical(numUnknown))
stopifnot(is.logical(numUnclass))
if (inherits(taxtab, "phyloseq")) {
tax <- as.matrix(taxtab@tax_table@.Data)
} else {
tax <- as.matrix(taxtab)
}
# Input check for numUnclass and numDupli
if (numUnclass) {
if (!grepl("<name>", numUnclassPat)) {
stop('Argument "numUnclassPat" must contain a space holder "<name>".')
}
if (!grepl("<num>", numUnclassPat)) {
stop('Argument "numUnclassPat" must contain a space holder "<num>".')
}
}
if (!is.null(numDupli)) {
if (!all(numDupli %in% colnames(tax))) {
stop('Ranks given with "numDupli" must match column names of ' ,
'taxonomic table.')
}
if (!grepl("<name>", numDupliPat)) {
stop('Argument "numDupliPat" must contain a space holder "<name>".')
}
if (!grepl("<num>", numDupliPat)) {
stop('Argument "numDupliPat" must contain a space holder "<num>".')
}
}
# Remove columns to ignore
if (!is.null(ignoreCols)) {
if (is.character(ignoreCols)) {
ignoreCols <- which(colnames(tax) %in% ignoreCols)
}
} else {
# Ignore columns that contain NAs only
ignore.tmp <- which(apply(tax, 2, function(x) all(is.na(x))))
if (length(ignore.tmp) == 1) {
message(paste0("Column ", ignore.tmp,
" contains NAs only and is ignored."))
ignoreCols <- ignore.tmp
} else if (length(ignore.tmp) > 1) {
message(paste0("Columns ", ignore.tmp,
" contain NAs only and are ignored."))
ignoreCols <- ignore.tmp
}
}
tax_orig <- tax
if (!is.null(ignoreCols)) {
tax <- tax_orig[, -ignoreCols]
}
# Define ranks
nranks <- ncol(tax)
ranks.tmp <- c("Life", "Domain", "Kingdom", "Phylum", "Class", "Order",
"Family", "Genus", "Species")
ranksAbb.tmp <- substr(tolower(ranks.tmp), 1, 1)
missRanks <- missRanksAbb <- FALSE
if (is.null(ranks)) {
if (any(ranks.tmp %in% colnames(tax))) {
ranks <- colnames(tax)
} else {
repl <- 0
ranks <- rep(NA, ncol(tax))
for (r in seq_along(ranks.tmp)) {
sel <- grep(paste0(ranksAbb.tmp, "_")[r], t(tax)) %% nranks
if (length(sel) != 0) {
sel <- sel[1]
if (sel == 0) sel <- nranks
ranks[sel] <- ranks.tmp[r]
repl <- repl + 1
}
}
if (repl != nranks) {
missRanks <- TRUE
}
}
} else {
if (length(ranks) != nranks) {
stop("Length of 'ranks' does not match number of columns")
}
}
if (is.null(ranksAbb)) {
if (missRanks) {
missRanksAbb <- TRUE
} else {
ranksAbb <- substr(tolower(ranks), 1, 1)
}
} else {
if (length(ranksAbb) != nranks) {
stop("Length of 'ranksAbb' does not match number of columns")
}
}
# Stop if full rank names are not given but needed
if (missRanks && (grepl("<rank>", pat) |
grepl("<Rank>", pat))) {
stop("Ranks could not be determined but are needed for the given pattern. ",
"Please provide 'ranks'")
}
if (missRanks && (grepl("<rank>", substPat) |
grepl("<Rank>", substPat) |
grepl("<subst_rank>", substPat) |
grepl("<subst_Rank>", substPat))) {
stop("Ranks could not be determined but are needed for the given ",
"substitute pattern. Please provide 'ranks'")
}
# Stop if abbreviated rank names are not given but needed
if (missRanksAbb && (grepl("<r>", pat) |
grepl("<R>", pat))) {
stop("Abbreviated ranks could not be determined but are needed for the ",
"given pattern. Please provide 'ranks' or 'ranksAbb'.")
}
if (missRanksAbb && (grepl("<r>", substPat) |
grepl("<R>", substPat) |
grepl("<subst_r>", substPat) |
grepl("<subst_R>", substPat))) {
stop("Ranks could not be determined but are needed for the given ",
"substitute pattern. Please provide 'ranks' or 'ranksAbb'.")
}
for (r in seq_along(ranksAbb.tmp)) {
tax <- gsub(paste0(ranksAbb.tmp[r], "__"), "", tax)
tax <- gsub(paste0(ranksAbb.tmp[r], "_"), "", tax)
}
#-----------------------------------------------------------------------------
### Make labels unique
unknownMat <- matrix(data = FALSE, nrow = nrow(tax), ncol = nranks)
dimnames(unknownMat) <- dimnames(tax)
# Add number to unclassified taxa
if (!is.null(unclass)) {
if (numUnclass) {
# Split pattern
pat.tmp <- unlist(strsplit(numUnclassPat, "<"))
pat.tmp <- unlist(strsplit(pat.tmp, ">"))
numUnclassPat <- pat.tmp
for (c in 1:nranks) {
# Detect unclassified taxa
isunclass <- tax[, c] %in% unclass
if (sum(isunclass) > 0) {
# Replace names according to pattern
tmp <- matrix(rep(numUnclassPat,
each = sum(isunclass)), nrow = sum(isunclass))
tmp[tmp == "name"] <- tax[isunclass, c]
tmp[tmp == "num"] <- 1:sum(isunclass)
tax[isunclass, c] <- apply(tmp, 1, paste, collapse="")
unknownMat[isunclass, c] <- TRUE
}
}
}
} else {
if (any(grepl("unclass", tax, ignore.case = TRUE))) {
unclassname <- unique(tax[grep("unclass", tax, ignore.case = TRUE)])
if (!all(unclassname %in% unknown)) {
unclassname <- unclassname[!unclassname %in% unknown]
warning(paste0('Taxonomic table contains unclassified taxa. ',
'Consider adding "', unclassname,
'" to argument "unknown".'))
}
}
}
# Replace unknowns by number
if (numUnknown) {
for (c in 1:nranks) {
# Add a number if the rank is unknown
isunknown <- tax[, c] %in% unknown
if (sum(isunknown) > 0) {
tax[isunknown, c] <- 1:sum(isunknown)
unknownMat[isunknown, c] <- TRUE
}
}
}
# Make duplicates unique
if (!is.null(numDupli)) {
# Split pattern
pat.tmp <- unlist(strsplit(numDupliPat, "<"))
pat.tmp <- unlist(strsplit(pat.tmp, ">"))
numDupliPat <- pat.tmp
for (c in seq_along(numDupli)) {
tax.tmp <- tax[, numDupli[c]]
dupli <- table(tax.tmp)[table(tax.tmp) > 1]
if (length(dupli) > 0) {
for (d in seq_along(dupli)) {
# Replace names according to pattern
tmp <- matrix(rep(numDupliPat, each = dupli[d]), nrow = dupli[d])
tmp[tmp == "name"] <- names(dupli[d])
tmp[tmp == "num"] <- 1:dupli[d]
tax.tmp[tax.tmp == names(dupli[d])] <- apply(tmp, 1, paste,
collapse="")
}
tax[, numDupli[c]] <- tax.tmp
}
}
}
#-----------------------------------------------------------------------------
pat.tmp <- unlist(strsplit(pat, "<"))
pat.tmp <- unlist(strsplit(pat.tmp, ">"))
pat <- pat.tmp
pat.tmp <- unlist(strsplit(substPat, "<"))
pat.tmp <- unlist(strsplit(pat.tmp, ">"))
substPat <- pat.tmp
for (i in 1:nrow(tax)) {
for (r in nranks:1) {
name <- tax[i, r]
isunknown <- unknownMat[i, r]
usesubst <- NA
if (isunknown) {
usesubst <- TRUE
# Which ranks are known?
knownr <- which(!unknownMat[i, ])
if (length(knownr) == 0) {
# If all ranks are unknown, take name of the highest rank
sr <- 1
if (r == 1) usesubst <- FALSE
} else {
# Select last known rank
sr <- knownr[length(knownr)]
}
} else {
usesubst <- FALSE
}
# Assign the right pattern
if (usesubst) {
patout <- substPat
} else {
patout <- pat
}
patout[patout == "r"] <- tolower(ranksAbb[r])
patout[patout == "R"] <- toupper(ranksAbb[r])
patout[patout == "rank"] <- tolower(ranks[r])
patout[patout == "Rank"] <- paste0(toupper(substr(ranks[r], 1, 1)),
substr(ranks[r], 2,
nchar(ranks[r])))
patout[patout == "name"] <- name
if (usesubst) {
patout[patout == "subst_r"] <- tolower(ranksAbb[sr])
patout[patout == "subst_R"] <- toupper(ranksAbb[sr])
patout[patout == "subst_rank"] <- tolower(ranks[sr])
patout[patout == "subst_Rank"] <- paste0(toupper(substr(ranks[sr], 1, 1)),
substr(ranks[sr], 2,
nchar(ranks[sr])))
patout[patout == "subst_name"] <- tax[i, sr]
}
taxname <- paste(patout, collapse = "")
tax[i, r] <- taxname
}
}
if (!is.null(ignoreCols)) {
taxout <- tax_orig
taxout[, (1:ncol(tax_orig))[-ignoreCols]] <- tax
} else {
taxout <- tax
}
if (inherits(taxtab, "phyloseq")) {
phyloseq::tax_table(taxtab) <- phyloseq::tax_table(taxout)
taxout <- taxtab
}
return(taxout)
}
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.