t4f: Predict taxonomic functional content via tax4fun

Description

Given a taxonomic abundance table prepared with the Silva reference database, predicts the functional content using a KO precalculated mapping table that maps the taxonomic abundance for a given tax_table to functional abundance content across a set of functional genes.

Usage

t4f(
  otu_table,
  rows_are_taxa,
  tax_table,
  reference_path,
  type = c("uproc", "pauda"),
  short = TRUE,
  cn_normalize = FALSE,
  sample_normalize = FALSE,
  scalar,
  drop = TRUE,
  verbose = FALSE
)

Arguments

otu_table

(required) Matrix or dataframe containing taxa abundances (counts, non-negative integers) across samples. Rows and columns must be uniquely named.

rows_are_taxa

(required) Logical flag indicating whether otu_table rows correspond to taxa (TRUE) or samples (FALSE).

tax_table

Matrix or dataframe containing Silva taxonimic information with row or column names corresponding to the otu_table. Silva species information is required.

reference_path

Folder path of the silva-to-kegg mapping file (t4f_silva_to_kegg.rds) and reference profiles (t4f_ref_profiles.rds). Must not be renamed.

type

Type of protein domain classification methods used to generate references (uproc or pauda). Defaults to uproc.

short

Logical flag whether to use a short or long read references. Defaults to TRUE.

cn_normalize

Logical flag for performing 16S rRNA copy number normalization. Defaults to FALSE.

sample_normalize

Logical flag to normalize functional predictions by the total functional abundance in a sample. Defaults to FALSE.

scalar

Value for scaling the topics over functions distrubution to predicted counts.

drop

Logical flag to drop empty gene columns after prediction. Defaults to TRUE.

verbose

Logical flag to print progress information. Defaults to FALSE.

Value

A list containing

fxn_table: A matrix of gene counts across topics.
fxn_meta: A list of functional metadata corresponding to fxn_table.
method_meta: A matrix of method specific metadata (FTU).

References

ABhauer, K. P., Wemheuer, B. Daniel, R., and Meinicke, P. (2015). Bioinformatics, 1-3. 31(17).

Examples

Run this code

# NOT RUN {
download_ref(destination='/references',reference='silva_ko')
predicted_functions <- t4f(otu_table=DAVID$OTU,rows_are_taxa=FALSE,
                           tax_table=DAVID$TAX,reference='/references',
                           type='uproc',short=TRUE,cn_normalize=TRUE,
                           sample_normalize=FALSE,scalar=NULL,drop=TRUE)
                           
# }
# NOT RUN {
# }