Skip to contents

Decontaminate a Phyloseq Object Using Specified Methods

Source: R/decontam.R
decontam.Rd

This function removes contamination from a phyloseq object using the decontam package. It supports the frequency, prevalence, or both methods for contaminant identification. Note that the prevalence method relies on the presence of blank samples; if blank samples are not available, only the frequency method can be performed.

Usage

decontam(
  physeq = resolved_tree_physeq,
  decon_method = c("frequency", "prevalence", "both"),
  blank = TRUE
)

Arguments

physeq

A phyloseq object containing the microbiome data. This is the input object that the function processes.

decon_method

A character string specifying the contamination removal method. Possible values are:

  • frequency: Identifies contaminants by examining the distribution of sequence feature frequencies as a function of the input DNA concentration.

  • prevalence: Identifies contaminants by comparing the prevalence (presence/absence across samples) of sequence features in true samples versus negative controls (blanks).

  • both: Applies both frequency and prevalence methods sequentially. Taxa flagged by either method are considered contaminants. This option requires that blank samples are available.

blank

A logical value indicating whether blank samples were included in the dataset.

  • TRUE: Blank samples are present, allowing the use of the both method.

  • FALSE: No blank samples are available; in this case, only the frequency method can be applied.

Value

A phyloseq object with contaminants removed. The decontaminated object is saved as an RDS file named <project_name>_phyloseq_asv_level_decontam.rds in the output_data/rds_files/Before_cleaning_rds_files directory.

Details

The function uses the decontam package to remove contaminants based on the specified method. If both is chosen, the function applies the frequency method first and then the prevalence method, flagging any taxa identified by either method as contaminants. In addition, diagnostic plots (showing read counts and contaminant prevalence) are generated and saved as PDF files.

Examples

if (FALSE) { # \dontrun{
# Example usage:
decontam(physeq = phyloseq_data, decon_method = "both", blank = TRUE)
} # }