subsetORFs {SQMtools}

Select ORFs

Description

Create a SQM object containing only the requested ORFs, and the contigs and bins that contain them. Internally, all the other subset functions in this package end up calling subsetORFs to do the work for them.

Usage

subsetORFs(
  SQM,
  orfs,
  tax_source = "orfs",
  trusted_functions_only = FALSE,
  ignore_unclassified_functions = FALSE,
  rescale_tpm = FALSE,
  rescale_copy_number = FALSE,
  contigs_override = NULL
)

Arguments

SQM

SQM object to be subsetted.

orfs

character. Vector of ORFs to be selected.

tax_source

character. Features used for calculating aggregated abundances at the different taxonomic ranks. Either "orfs" or "contigs" (default "orfs").

trusted_functions_only

logical. If TRUE, only highly trusted functional annotations (best hit + best average) will be considered when generating aggregated function tables. If FALSE, best hit annotations will be used (default FALSE).

ignore_unclassified_functions

logical. If FALSE, ORFs with no functional classification will be aggregated together into an "Unclassified" category. If TRUE, they will be ignored (default FALSE).

rescale_tpm

logical. If TRUE, TPMs for KEGGs, COGs, and PFAMs will be recalculated (so that the TPMs in the subset actually add up to 1 million). Otherwise, per-function TPMs will be calculated by aggregating the TPMs of the ORFs annotated with that function, and will thus keep the scaling present in the parent object (default FALSE).

rescale_copy_number

logical. If TRUE, copy numbers with be recalculated using the RecA/RadA coverages in the subset. Otherwise, RecA/RadA coverages will be taken from the parent object. By default it is set to FALSE, which means that the returned copy numbers for each function will represent the average copy number of that function per genome in the parent object.

contigs_override

character. Optional vector of contigs to be included in the subsetted object.

Value

SQM object containing the requested ORFs.

A note on contig/bins subsetting

While this function selects the contigs and bins that contain the desired orfs, it DOES NOT recalculate contig/bin abundance and statistics based on the selected ORFs only. This means that the abundances presented in tables such as SQM$contig$abund or SQM$bins$tpm will still refer to the complete contigs and bins, regardless of whether only a fraction of their ORFs are actually present in the returned SQM object. This is also true for the statistics presented in SQM$contigs$table and SQM$bins$table.

Examples

data(Hadza)
# Select the 100 most abundant ORFs in our dataset.
mostAbundantORFnames = names(sort(rowSums(Hadza$orfs$tpm), decreasing=TRUE))[1:100]
mostAbundantORFs = subsetORFs(Hadza, mostAbundantORFnames)