compute_prior_prob: Compute prior probabilities for putative theoretical features

In szymczak-lab/preprocessHighResMS: Preprocessing of spectra data from high resolution mass spectrometry

Description

Estimates prior probabilities of all possible theoretical features for each measured feature.

Usage

compute_prior_prob(
  hits.m,
  info.features,
  dat,
  pk = NULL,
  ppm,
  ppm.unknown = NULL,
  pr.limit = 1e-05
)

Arguments

hits.m	matrix with information about overlapping theoretical features (e.g. as generated by the function find_hits).
info.features	data.frame with information about theoretical possible features (e.g. as generated by the function chem_formula_2_adducts).
dat	data.frame with information about measured features with columns id (= unique identifier), mz (= measured mz value), intensity (= measured intensity).
pk	Numeric vector with probabilities of initial confidence of the presence of each theoretical feature.
ppm	Numeric. Resolution of mass spectrometer.
ppm.unknown	Numeric. Resolution assigned to unknown theoretical features (i.e. not contained in info.features).
pr.limit	Numeric. Lowest probability not set to 0 (default = 1e-05).

Value

matrix with measured features in rows and theoretical possible features in columns. The cell i,j contains the prior probability that measured feature i belongs to theoretical feature j.

Examples

data("se.example")
data("info.features")

dat = prepare_data_for_annotation(se = se.example)
hits.m = find_hits(info.features = info.features,
                   dat = dat,
                   ppm = 20)

prior.prob = compute_prior_prob(hits.m = hits.m,
                                info.features = info.features,
                                dat = dat,
                                ppm = 20)

szymczak-lab/preprocessHighResMS documentation built on Oct. 6, 2020, 12:50 a.m.