spiky: Spike-in calibration for cell-free MeDIP

spike_counts

R Documentation

use the index of a spiked BAM/CRAM file for spike contig coverage

Description

It dawned on me one day that we don't even have to bother reading the file if we have an index for a spiked BAM/CRAM result, since any fragments that map properly to the spike contigs are generated from synthetic templates. This function takes an index and a spike database (usually a DataFrame) as inputs and provides a rough coverage estimate over "rehabilitated" contig names (i.e., canonicalized contigs mapping to the database) as its output.

Usage

spike_counts(
  bam,
  spike,
  sep = "_",
  ref = "spike",
  verbose = FALSE,
  dump_idx = FALSE
)

Arguments

`bam`	the BAM or CRAM file (MUST HAVE AN INDEX)
`spike`	a data.frame, DataFrame, or similar with spikes
`sep`	separator character in contig names ("_")
`ref`	reference name for spike genome ("spike")
`verbose`	be verbose? (FALSE)
`dump_idx`	dump the renamed idxstats to aggregate? (FALSE)

Details

The argument spike has no default since we are attempting to refactor the spike-in databases into their own data packages and allow more general use.

Value

               a GRanges of spike-in contig read counts

Examples

data(spike, package="spiky")
sb <- system.file("extdata", "example.spike.bam", package="spiky",
                  mustWork=TRUE) 
spike_counts(sb, spike=spike)

trichelab/spiky documentation built on Sept. 17, 2022, 8:44 a.m.