Normalise DRAGEN coverage summary-metric names

Rewrites the variable column of the summary rows of a parsed DRAGEN *_coverage_metrics.csv into region-agnostic names, so a single schema serves every coverage region. Two rewrites are applied:

• over <region> suffixes (e.g. Average alignment coverage over genome) are stripped;

• in <region> suffixes (e.g. Aligned bases in genome) collapse to in region, keeping them distinct from the plain Aligned bases total.

The coverage-bin rows (PCT of <region> with coverage [lo: hi)) are split off into the separate bins table by DragenCov and parsed with dragen_cov_bin_split(), so they never pass through this function. The region itself is carried in the output prefix (via refine_files()), so unlike the variant-caller metrics no region column is emitted.

Usage

dragen_cov_metric_normalize(v)

Arguments

v

(character())
Vector of raw metric names.

Value

(character()) Normalised metric names.

Examples

dragen_cov_metric_normalize(c(
  "Aligned bases",
  "Aligned bases in genome",
  "Average alignment coverage over QC coverage region"
))
#> [1] "Aligned bases"              "Aligned bases in region"   
#> [3] "Average alignment coverage"