Rename miRNA features in a matrix or named vector

Convenience wrapper around resolve_mirna_aliases that accepts a samples × features matrix or a named numeric vector and renames the features (column names of the matrix, or names() of the vector) to the target namespace. Unresolved features are either dropped (keep_unresolved = FALSE) or retained with their original name (keep_unresolved = TRUE, default).

Usage

apply_mirna_aliases(
  x,
  target_namespace = c("mirna_name", "mimat"),
  table = mirna_alias_table(),
  keep_unresolved = TRUE,
  verbose = FALSE,
  ...
)

Arguments

x

A samples × features matrix with column names, or a named numeric vector.

target_namespace

One of "mirna_name" (default) or "mimat". Passed to resolve_mirna_aliases.

table

Alias table. Defaults to mirna_alias_table().

keep_unresolved

Logical. If TRUE (default), features whose names cannot be resolved are kept with their original name. If FALSE, unresolved features are dropped from the output.

verbose

Logical. Passed to resolve_mirna_aliases.

...

Additional arguments passed to resolve_mirna_aliases.

Value

An object of the same class and structure as x with feature names mapped to the target namespace. When keep_unresolved = FALSE and some features are unresolved, those columns / elements are removed.

References

Kozomara A., Birgaoanu M., Griffiths-Jones S. (2019) miRBase: from microRNA sequences to function. Nucleic Acids Research 47(D1): D155–D162. DOI: doi:10.1093/nar/gky1141

Examples

# Named numeric vector
v <- c(MIMAT0001631 = 12.3, MIMAT0000062 = 4.1, unknown_probe = 0.9)
apply_mirna_aliases(v)
#>  hsa-miR-451a hsa-let-7a-5p unknown_probe 
#>          12.3           4.1           0.9 

# Matrix
M <- matrix(runif(6), nrow = 2,
            dimnames = list(c("S1", "S2"),
                            c("MIMAT0001631", "MIMAT0000062", "junk")))
apply_mirna_aliases(M, keep_unresolved = FALSE)
#>    hsa-miR-451a hsa-let-7a-5p
#> S1    0.0539110    0.02560094
#> S2    0.9550958    0.92076314