phyloseq_filter_sample_wise_abund_trim: Filter rare OTUs based on minimum abundance threshold.

In vmikk/metagMisc: Miscellaneous functions for metagenomic analysis

Filter rare OTUs based on minimum abundance threshold.

Description

This function performs sample-wise OTU abundance trimming.

Usage

phyloseq_filter_sample_wise_abund_trim(
  physeq,
  minabund = 10,
  relabund = FALSE,
  rm_zero_OTUs = TRUE
)

Arguments

physeq	A phyloseq-class object
minabund	Abundance threshold (default, 10)
relabund	Logical; perform trimming based on relative abundances (default, FALSE)
rm_zero_OTUs	Logical, remove OTUs with zero total abundance

Details

OTUs can be considered as rare if they comprise fewer than X (e.g., 10) sequences within a sample. This function is intented to censore OTU abundance (unsing an arbitrary threshold) on a sample-wise basis.

Trimming can be performed based on relative abundances of OTUs within a sample ('relabund = TRUE'), but the orginal OTU count will be returned. For this purpose 'minabund' parameter should be provided in a range of (0,1] (e.g., use 'minabund = 0.1, relabund = TRUE' to remove OTUs with relative abundance < 10

Value

Phyloseq object with a filtered data.

Examples

# Load data
data(GlobalPatterns)

# Trim GlobalPatterns data (19216 taxa) by removing OTUs with less that 10 reads
GP1 <- phyloseq_filter_sample_wise_abund_trim(GlobalPatterns, minabund = 10) # 10605 taxa

# Trim GlobalPatterns data by removing OTUs with relative abundance less than 1%
GP2 <- phyloseq_filter_sample_wise_abund_trim(GlobalPatterns, minabund = 0.01, relabund = TRUE) # 258 taxa

# Compare raw and trimmed data
phyloseq_summary(GlobalPatterns, GP1, GP2, cols = c("GlobalPatterns", "Trimmed 10 reads", "Trimmed 1 percent"))

vmikk/metagMisc documentation built on Feb. 14, 2024, 2:29 a.m.