metamicrobiomeR (version 1.2)

pathway.compare: Compare (kegg) pathway abundance

Description

This is a slightly modified version of the taxa.compare function. It compares pathway abundance generated from PICRUSt analysis between groups using different methods (apply to pathway summary tables already merged to mapping file and put in a list (e.g.level 1, 2 and 3)). Specifically, it compares relative abundances of bacterial functional pathways at all levels using GAMLSS or LM/LMEM and compares of log(absolute abundances) of bacterial functional pathways at all levels using LM/LMEM.

Usage

pathway.compare(
  pathtab,
  mapfile,
  sampleid = "sampleid",
  pathsum = "rel",
  stat.med = "gamlss",
  transform = "none",
  comvar,
  adjustvar,
  personid = "personid",
  longitudinal = "yes",
  p.adjust.method = "fdr",
  percent.filter = 0.05,
  relabund.filter = 5e-05,
  pooldata = FALSE
)

Arguments

pathtab

list of pathway abundance table of all levels.

mapfile

mapping file or file containing covariates.

sampleid

variable containing sample id to be matched between pathway abundance table and mapping file.

pathsum

type of abundance to be compared. Options are "rel" for relative abundance or "log" for log of absolute abundance. Default is "rel".

stat.med

statistical method for comparison. stat.med can be "lm" for LM/LMEM (usable for both pathsum="rel" or "log") or "gamlss" for GAMLSS with BEZI family (gamlss only make sense if pathsum="rel" ).

transform

transformation of relative abundance data. Options are "none" for no transformation, "asin.sqrt" for arcsine transformation, "logit" for logit transformation. Default is "none".

comvar

main variable for comparison.

adjustvar

variables to be adjusted.

personid

name of variable for person id in mapping file (applicable for longitudinal data)

longitudinal

whether the data is longitudinal. Default is "yes".

p.adjust.method

method for multiple testing adjustment. Available options are those of the p.adjust function. Default is "fdr".

percent.filter

prevalence threshold (the percentage of number of samples the taxa/pathway available). Default is 0.05.

relabund.filter

relative abundance threshold (the minimum of the average relative abundance for a taxa/pathway to be retained). Default is 0.00005.

pooldata

whether the data is pooled from multiple studies. Default is FALSE.

Value

matrice of coefficients, standard errors, p-values and multiple testing adjusted p-values of all variables in the models.

Examples

# NOT RUN {
#Load Bangladesh pathway and metadata
data(kegg.12)
data(covar.rm)
# Comparison of pathway relative abundances for some first pathways of level 1 only
# and assuming crosssectional data (to save running time)
path1<-pathway.compare(pathtab=list(kegg.12[[1]][, 1:2]),
mapfile=covar.rm,sampleid="sampleid",pathsum="rel", stat.med="gamlss",
comvar="gender",adjustvar=c("age.sample","bf"), longitudinal="no",
p.adjust.method="fdr", percent.filter=0.05,relabund.filter=0.00005)
taxcomtab.show(taxcomtab=path1$l1, sumvar="path",tax.lev="l2",
tax.select="none", showvar="genderMale", p.adjust.method="fdr",p.cutoff=1)
# }