clean_profiles: Preprocess a co-elution profile matrix

In fosterlab/PrInCE: Predicting Interactomes from Co-Elution

Description

Clean a matrix of co-elution/co-fractionation profiles by (1) imputing single missing values with the average of neighboring values, (2) replacing missing values with random, near-zero noise, and (3) smoothing with a moving average filter.

Usage

clean_profiles(
  profile_matrix,
  impute_NA = TRUE,
  smooth = TRUE,
  smooth_width = 4,
  noise_floor = 0.001
)

Arguments

profile_matrix	a numeric matrix of co-elution profiles, with proteins in rows, or a MSnSet object
impute_NA	if true, impute single missing values with the average of neighboring values
smooth	if true, smooth the chromatogram with a moving average filter
smooth_width	width of the moving average filter, in fractions
noise_floor	mean value of the near-zero noise to add

Value

a cleaned matrix

Examples

data(scott)
mat <- scott[c(1, 16), ]
mat_clean <- clean_profiles(mat)

fosterlab/PrInCE documentation built on Dec. 13, 2020, 5:50 a.m.