filter_peaks_by_rsd: Filter features by RSD% of QC samples

In pmp: Peak Matrix Processing and signal batch correction for metabolomics datasets

Description

Metabolomics datasets often contain 'features' with irreproducible peak intensity values, or with large numbers of missing values. This tool facilitates the remove of such features from a data matrix, based upon relative standard deviation of intensity values for a given feature within specified QC samples.

Usage

filter_peaks_by_rsd(df, max_rsd, classes, qc_label, remove_peaks = TRUE)

Arguments

df	A matrix-like (e.g. an ordinary matrix, a data frame) or RangedSummarizedExperiment-class object with all values of class numeric() or integer() of peak intensities, areas or other quantitative characteristic.
max_rsd	numeric(), threshold of QC RSD% value
classes	character(), vector of class labels. Must be the same length as the number of sample in the input peak table. If input is SummarizedExperiment object, use SummarizedExperiment_object$meta_data_column_name.
qc_label	character(1) or NULL, class label used to identify QC samples.
remove_peaks	logical(1), remove filtered features from peak matrix or not.

Value

Object of class SummarizedExperiment. If input data are a matrix-like (e.g. an ordinary matrix, a data frame) object, function returns numeric() matrix-like object of filtered data set. Function flags are added to the object attributes and is a DataFrame-class with five columns. The same DataFrame-class object containing flags is added to rowData() element of SummarizedExperiment object as well.

Columns in rowData() or flags element contain:
rsd_QC numeric(), RSD% value of QC samples per feature;
rsd_flags integer(),if 0 feature is flagged to be removed.

Examples

df <- MTBLS79[ ,MTBLS79$Batch == 1]
out <- filter_peaks_by_rsd(df=df, max_rsd=20,
    classes=df$Class, qc_label='QC')

pmp documentation built on April 1, 2021, 6:01 p.m.