ssvFetchSignal: signal loading framework

Description Usage Arguments Details Value Examples

View source: R/functions_fetch_signal.R

Description

Does nothing unless load_signal is overridden to carry out reading data from file_paths (likely via the appropriate ssvFetch* function, ie. ssvFetchBigwig or ssvFetchBam

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
ssvFetchSignal(
  file_paths,
  qgr,
  unique_names = NULL,
  names_variable = "sample",
  file_attribs = NULL,
  win_size = 50,
  win_method = c("sample", "summary")[1],
  return_data.table = FALSE,
  load_signal = function(f, nam, qgr) {     warning("nothing happened, ",
    "supply a function to", "load_signal parameter.") },
  n_cores = getOption("mc.cores", 1),
  n_region_splits = 1,
  force_skip_centerFix = FALSE
)

Arguments

file_paths

character vector of file_paths to load from. Alternatively, file_paths can be a data.frame or data.table whose first column is a character vector of paths and additial columns will be used as metadata.

qgr

GRanges of intervals to return from each file

unique_names

unique file ids for each file in file_paths. Default is names of file_paths vector

names_variable

character, variable name for column containing unique_names entries. Default is "sample"

file_attribs

optional data.frame/data.table with one row per item in file paths. Each column will be a variable added to final tidy output.

win_size

numeric/integer window size resolution to load signal at. Default is 50.

win_method

character. one of c("sample", "summary"). Determines if viewGRangesWinSample_dt or viewGRangesWinSummary_dt is used to represent each region in qgr.

return_data.table

logical. If TRUE data.table is returned instead of GRanges, the default.

load_signal

function taking f, nam, and qgr arguments. f is from file_paths, nam is from unique_names, and qgr is qgr. See details.

n_cores

integer number of cores to use. Uses mc.cores option if not supplied.

n_region_splits

integer number of splits to apply to qgr. The query GRanges will be split into this many roughly equal parts for increased parallelization. Default is 1, no split.

force_skip_centerFix

boolean, if TRUE all query ranges will be used "as is". This is already the case by default if win_method == "summary" but may have applications where win_method == "sample".

Details

load_signal is passed f, nam, and qgr and is executed in the environment where load_signal is defined. See ssvFetchBigwig and ssvFetchBam for examples.

Value

A GRanges with values read from file_paths at intervals of win_size. Originating file is coded by unique_names and assigned to column of name names_variable. Output is data.table is return_data.table is TRUE.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
library(GenomicRanges)
bam_f = system.file("extdata/test.bam",
    package = "seqsetvis", mustWork = TRUE)
bam_files = c("a" = bam_f, "b" = bam_f)
qgr = CTCF_in_10a_overlaps_gr[1:2]
qgr = resize(qgr, 500, "center")
load_bam = function(f, nam, qgr) {
    message("loading ", f, " ...")
    dt = seqsetvis:::ssvFetchBam.single(bam_f = f,
                      qgr = qgr,
                      win_size = 50,
                      fragLen = NULL,
                      target_strand = "*",
                      return_data.table = TRUE)
    dt[["sample"]] = nam
    message("finished loading ", nam, ".")
    dt
}
ssvFetchSignal(bam_files, qgr, load_signal = load_bam)

seqsetvis documentation built on Nov. 8, 2020, 5:57 p.m.