MSclassifR (version 0.4.0)

PredictFastClass: Prediction of the category to which a mass spectrum belongs using linear regressions of mass spectra.

Description

For each mass peak in a list of mass peaks, a linear regression is performed between the mass spectrum and mass spectra corresponding to a category. This is performed for each category and associated to an Akaike Information Criterium. Next, the AIC are used to determine the belonging of a mass spectrum to a category. It also provides a probability that the mass spectrum does not belong to any of the input categories.

Usage

PredictFastClass(peaks,
                 mod_peaks,
                 Y_mod_peaks,
                 moz="ALL",
                 tolerance = 6,
                 toleranceStep = 2,
                 normalizeFun = TRUE,
                 noMatch = 0)

Value

Returns a dataframe containing AIC criteria by category for each mass spectrum in peaks. The AIC criterion should be minimal for the most probable category. The pred_cat column is the predicted category for each mass spectrum in peaks. The p_not_in_DB is the minimal p-value of several Fisher tests testing if all the linear coefficients associated to mass spectra of a category are null. It can be interpreted as a p-value that the mass spectrum is not present in the input database.

Arguments

peaks

a list of MassPeaks objects (see MALDIquant R package).

mod_peaks

an intensity matrix corresponding to mass spectra for which the category is known. Each column is a mass-to-charge value, each row corresponds to a mass spectrum.

Y_mod_peaks

a factor with a length equal to the number of mass spectra in mod_peaks and containing the categories of each mass spectrum in mod_peaks.

moz

a vector with the set of shortlisted mass-to-charge values that corresponds to mass-to-charge values in the columns of mod_peaks. By default, all the mass-to-charge values in mod_peaks are used.

tolerance

a numeric value of accepted tolerance to match peaks to the set of shortlisted mass-to-charge values. It is fixed to 6 Da by default.

toleranceStep

a numeric value added to the tolerance parameter to match peaks to the set of shortlisted mass-to-charge values. It is fixed to 2 Da by default.

normalizeFun

a logical value, if TRUE (default) the maximum intensity will be equal to 1, the other intensities will be expressed in ratio to this maximum.

noMatch

a numeric value used to replace intensity values if there is no match detected between peaks and the set of shortlisted mass-to-charge values moz. It is fixed to 0 by default.

Details

Godmer et al. (2025) presents a comparison of different pipelines using PredictFastClass that can help you to optimize your workflow. For a comprehensive guide, additional applications, and detailed examples of using this package, please visit our GitHub repository: here.

References

Alexandre Godmer, Yahia Benzerara, Emmanuelle Varon, Nicolas Veziris, Karen Druart, Renaud Mozet, Mariette Matondo, Alexandra Aubry, Quentin Giai Gianetto, MSclassifR: An R package for supervised classification of mass spectra with machine learning methods, Expert Systems with Applications, Volume 294, 2025, 128796, ISSN 0957-4174, tools:::Rd_expr_doi("10.1016/j.eswa.2025.128796").

Examples

# \donttest{

library("MSclassifR")
library("MALDIquant")

# load mass spectra and their metadata
data("CitrobacterRKIspectra","CitrobacterRKImetadata", package = "MSclassifR")
# standard pre-processing of mass spectra
spectra <- SignalProcessing(CitrobacterRKIspectra)
# detection of peaks in pre-processed mass spectra
peaks <- peaks <- MSclassifR::PeakDetection(x = spectra, averageMassSpec=FALSE)
# matrix with intensities of peaks arranged in rows (each column is a mass-to-charge value)
IntMat <- MALDIquant::intensityMatrix(peaks)
rownames(IntMat) <- paste(CitrobacterRKImetadata$Strain_name_spot)
# remove missing values in the matrix
IntMat[is.na(IntMat)] <- 0
# normalize peaks according to the maximum intensity value for each mass spectrum
IntMat <- apply(IntMat,1,function(x) x/(max(x)))
# transpose the matrix for statistical analysis
X <- t(IntMat)
# define the known categories of mass spectra for the classification
Y <- factor(CitrobacterRKImetadata$Species)

#Predict species without peak selection using a tolerance of 1 Da
res = PredictFastClass(peaks=peaks[1:5],
                       mod_peaks=X,
                       Y_mod_peaks=Y,
                       tolerance = 1)

#comparing predicted categories (species) and the truth
cbind(res$pred_cat,as.character(Y[1:5]))

# The method can be applied after a peak selection step
a <- SelectionVar(X,
                  Y,
                  MethodSelection = c("RFERF"),
                  MethodValidation = c("cv"),
                  PreProcessing = c("center","scale","nzv","corr"),
                  NumberCV = 2,
                  Metric = "Kappa",
                  Sizes = c(20:40),
                  Sampling = "up")

#Predict species from selected peaks using a tolerance of 1 Da
res = PredictFastClass(peaks=peaks[1:5],
                       moz = a$sel_moz,
                       mod_peaks=X,
                       Y_mod_peaks=Y, tolerance = 1)

#comparing predicted categories (species) and the truth
cbind(res$pred_cat,as.character(Y[1:5]))

# }