Error in check_rank_names(ps) : ranks of taxonimic profile must be one of Kingdom, Phylum, Class, Order, Family, Genus, Species

[KatrinaLutap]

Dear @yiluheihei, thank you for this package.

I had this error message while using the package:
Error in check_rank_names(ps) :
ranks of taxonimic profile must be one of Kingdom, Phylum, Class, Order, Family, Genus, Species

I think it is because my phyloseq object has 8 taxonomic ranks based from my taxonomy database:

rank_names(ps)
[1] "Domain" "Subdomain" "Kingdom" "Phylum" "Subphylum" "Class" "Genus"
[8] "Species"

Does this mean I can not use more than 8 taxonomic ranks and is there a way to go around this?

Thank you.

Best,
Katrina

[Luke-ebbis]

Hey @KatrinaLutap and @yiluheihei I get the same error with this script. What may help to prevent the error?

#'@title Linear discriminant analysis Effect Size analysis script.
#'@description In this script the LDA effect sizes are investigated for: 


# Prepare the environment ----
library(tidyverse)
library(phyloseq)
library(microbiomeMarker)
library(microViz)
library(readxl)

source("utils/utils.R")


# Reading in the data ----

otu_mat <- read.csv("data/OTU_table_Exp2_24hrs.csv")
tax_mat <- read.csv("data/Taxa_Exp_1_2_Cleaned.csv")
samples_df <- read.csv("data/Metadata_Exp2_24hrs.csv")

# phy <- make_phyloseq(otu_mat, tax_mat, samples_df)

otu_mat <- otu_mat %>%
  tibble::column_to_rownames("asv")
tax_mat <- tax_mat %>%
  tibble::column_to_rownames("asv")
samples_df <- samples_df %>%
  tibble::column_to_rownames("samples")
otu_mat <- as.matrix(otu_mat)
tax_mat <- as.matrix(tax_mat)
OTU <- otu_table(otu_mat, taxa_are_rows = TRUE)
TAX <- tax_table(tax_mat)
samples <- sample_data(samples_df)
phyloseq_object_all <- phyloseq(OTU, TAX, samples)
phyloseq_object_all <- prune_taxa(taxa_sums(phyloseq_object_all) > 0,
                                  phyloseq_object_all)
message("Created the phyloseq object")


lef_out <- run_lefse(phyloseq_object_all,
                     taxa_rank = "none",
                     group = "Treatment", norm = "CPM",
                     kw_cutoff = 0.05, lda_cutoff = 2)
                     

[DebnathSanjit]

Dear @KatrinaLutap and @Luke-ebbis I'm also having a similar issue with my 18S (microeukaryotes) dataset with 9 taxonomic ranks (Domain, Supergroup, Division, Subdivision, Class, Order, Family, Genus, Species" ). Have you got any solution for this?

[Luke-ebbis]

Looking at the code, the function check_rank_names checks whether the names of the taxa are all in the variable available_ranks.

microbiomeMarker/R/utilities.R

Lines 31 to 36 in afe4745

	if (!all(ps_ranks %in% available_ranks)) {
	stop(
	"ranks of taxonimic profile must be one of ",
	paste(available_ranks, collapse = ", ")
	)
	}

This global constant is defined here:

microbiomeMarker/data-raw/available_ranks.R

Lines 1 to 11 in afe4745

	# available taxonomic ranks, Summarize represents summarized tax
	available_ranks <- c(
	"Kingdom", "Phylum", "Class", "Order",
	"Family", "Genus", "Species"
	)
	available_ranks <- factor(
	available_ranks,
	levels = available_ranks
	)
	usethis::use_data(available_ranks, internal = TRUE, overwrite = TRUE)

So in my view the one way to deal with this is to

1. Make sure your taxa names match available_ranks ranks exactly.
2. Modify the package in some way to allow more classes.