Skip to contents

Read a user-supplied system / module / gene csv as a reference bundle

Source: R/reference_csv.R
dmsa_reference_csv.Rd

One flat file, one row per gene-within-module (or per gene-within-system if there is no module layer). Column names are matched case- and punctuation-insensitively, so System, system_id and "System ID" are all accepted.

Usage

dmsa_reference_csv(
  path,
  name = NULL,
  version = NA_character_,
  notes = NULL,
  min_genes = 3L,
  quiet = FALSE
)

Arguments

path

Path to the csv (anything utils::read.csv accepts).

name, version, notes

Provenance for the bundle; name defaults to the file name.

min_genes

Systems with fewer genes than this are dropped with a message - a set of two genes is not a set test. Set to 1 to keep everything.

quiet

Suppress the summary message.

Value

A dmsa_reference.

Details

Required: a system column (system, system_id, pathway, set) and a gene column (gene, symbol, gene_symbol).

Optional:

  • module (module, subsystem, submodule, module_id) - adds the module level to the cascade.

  • w_g (wg, weight, polarity, sign) - the gene's effect on system activation, -1/0/+1 or continuous in [-1, 1]. Without it the bundle carries no polarity and only gene-level DMSA is valid.

  • anchor (anchor, is_anchor, driver) - TRUE for the genes that define what "more activation" means. Without it, anchors are taken from role == "driver" if a role column exists, else inferred from the most positive w_g with a warning.

  • role, confidence, source - carried through untouched.

Examples

f <- tempfile(fileext = ".csv")
dmsa_reference_template(f)
#> template written to /var/folders/yd/5vhpvlls31x18db51sk7y_zh0000gn/T//RtmpUh53Rl/file76646764e2e7.csv
#>   system + gene are required; module, w_g, anchor, role, confidence, source are optional.
#>   w_g: +1 raises system activation, -1 opposes it, 0 off-axis; continuous values in [-1, 1] are allowed.
#>   anchor: TRUE for the genes that DEFINE activation for that system.
ref <- dmsa_reference_csv(f)
#> dmsa reference: file76646764e2e7.csv
#>   8 genes across 2 systems, 6 modules
#>   polarity: 8 gene-system pairs, continuous (6 activating, 2 braking, 0 off-axis)
#>   anchors: user (2 genes)