R/taxacolor.R

In ravinpoudel/PhONA: PhONA provides a framework to select a testable and manageable number of OTUs/ASVs to support microbiome-based agriculture

#' Function to assign color to the phyloseq object
#'
#' This function takes in phyloseq object, association matix, p value matrix
#' and create a combined OTU-OTU and OTU-Phenotype network. User can select model to
#' define OTU-Phontype assocaition.
#' @param phyobj A phyloseq object which combined OTU count, taxonomy and metadata
#' @param coloredby Select a taxonomic rank for coloring nodes in the network diagram. Possible option: "Kingdom","Phylum","Class","Order","Family","Genus","Species"
#' @return phyobj2 An updated phyloseq object with color information
#' @export


taxacolor <- function(phyobj, coloredby="x"){

  possible_tax <- c("Kingdom","Phylum","Class","Order","Family","Genus","Species")

  if (coloredby %in% possible_tax) {
    # create a color palltet using coloredby
    # number of unique color needed

    tc_tdata = data.frame(as(tax_table(phyobj), "matrix"))

    mod.vec <- tc_tdata %>%
      pull(coloredby) %>%
      as.character()
    n <- length(unique(mod.vec))
    mod.num <- as.numeric(as.factor(mod.vec))
    ucolors <- distinctColorPalette(n)

    # we need to pass the variable by reference
    # https://stackoverflow.com/questions/29190455/how-to-pass-parameters-by-reference-in-r
    eval.parent(substitute(colmn <-coloredby))

    getcolmn <- paste0(colmn, "_color")

    tc_tdata[getcolmn] <- sapply(mod.num, function(j) ucolors[j])
    # keep back as phyobj and return

    tax.mat <- tax_table(as.matrix(tc_tdata))

    otu.mat = otu_table(phyobj)
    sample.mat <- sample_data(phyobj)

    phyobj2 = phyloseq(otu.mat, tax.mat, sample.mat)
    return(phyobj2)
  } else {
    cat(" Provided coloredby: ", coloredby,"\n")

    stop("Provide correct coloredby paramter. Available options are: Kingdom,
         Phylum, Class, Order, Family, Genus, Species")
    }
}