metaGEENOME is a 16S rRNA metagenomic analysis tool encompassing nearly all downstream analysis steps. These steps include preprocessing to filter zero-inflated and low-abundance data, exploring the dataset through various plots, calculating alpha and beta diversity, generating ordination plots, testing null hypotheses, and applying a novel differential expression method using generalized estimating equations. As mentioned earlier, the results of all the steps are compiled into a single, detailed PDF file with a well-organized folder structure.
# The required packages list:
list.of.packages <- c("phyloseq", "microbiome","remotes")
new.packages <- list.of.packages[!(list.of.packages %in% installed.packages()[,"Package"])]
if(length(new.packages)) install.packages(new.packages)
# Load all required packages and show version
for(i in list.of.packages)
{
print(i)
print(packageVersion(i))
library(i, quietly=TRUE, verbose=FALSE, warn.conflicts=FALSE, character.only=TRUE)
}
if (!requireNamespace("metaGEENOME", quietly = TRUE)) {
# Install "metaGEENOME" from GitHub
remotes::install_github("Ahmed-A-Mohamed/metaGEENOME")
}
- you can read the detailed of metaGEENOME package from help in r
library(metaGEENOME)
?metaGEENOME
- call phyloseq package on R
library(phyloseq)
Important
Make sure that the data has the same format as phyloseq before importing.
This image is taken from phyloseq manual.
# load A two-week diet swap study between western (USA) and traditional (rural Africa) diets (Lahti et al. 2014).
data(dietswap, package = "microbiome")
phyloseq_data <- dietswap
# call library metaGEENOME
library(metaGEENOME)
# detect various parameters
# enter your phyloseq_data variable to be named "physeq"
physeq = phyloseq_data
# make sure from the column names in sample_data(physeq)
variables <- c("bmi_group","timepoint.within.group") # variables & cofounders for RCM
color_label <- c("bmi_group") # is the interested variable & in RCM
shape_label <- c("sex") # in RCM only
id = "subject" # individual id (personal id for each participant even if repeated measures in time series)
sample_var = "sample" # sample ID (equal to rownames of physeq@sam_data)
# Apply GEE downstream optionality if you don't need to re-analysis your data.
GEE_analysis <- TRUE
model_form <- c("bmi_group","timepoint.within.group")
# your need to make sure that phyloseq contain tax_table() before applying alpha & beta diversity
AlphaBetaDiversity = TRUE
BetaDiversity.distance <- "bray" # choose from these distances => distanceMethodList https://joey711.github.io/phyloseq/distance.html
fill_Alpha = "Family" # choose your interested taxonomic level (Please make sure tax_table() in physeq has that's level)
axes = c(1,2) # You need to detect the axes before plotting the ordination after scree plot
permanova.distance <- c("bray") # https://rdrr.io/bioc/phyloseq/man/distanceMethodList.html
Permanova.strata = NULL # add strata if you need
permanova.permutation_number <- 9999
# Preprocessing parameters
group_var = NULL
out_cut = 0.05
zero_cut = 0.9
lib_cut = 1000
neg_lb = FALSE
adj_pvalue <- "BH" # choose from these => c("holm", "hochberg", "hommel", "bonferroni", "BH", "BY", "fdr", "none")
Warning
Ensure that you are in the R directory before running metaGEENOME because the results will be applied in that directory.
# get your working directory
getwd()
# Run metaGEENOME
res <- metaGEENOME(physeq, variables, id, sample_var, group_var, out_cut, zero_cut,
lib_cut, neg_lb, model_form, alpha, n_cl, prv_cut,AlphaBetaDiversity,
color_label,BetaDiversity.distance,shape_label,axes,permanova.distance,
permanova.strata,permanova.permutation_number,adj_pvalue,GEE_analysis,fill_Alpha)
======= metaGEENOME is a 16S rRNA metagenomic analysis tool that encompasses nearly all steps of downstream analysis. These steps include preprocessing to filter zero-inflated and low abundance data, exploring the dataset through various plots, calculating alpha and beta diversity, generating ordination plots, testing null hypotheses, and applying a novel differential expression method using generalized estimating equations. The results of all the aforementioned steps are compiled into a single, detailed PDF file within a well-organized folder structure.
3b5f752 (Initial commit)