FeatureSelection: High-quality feature selection

In Waddlessss/MAFFIN: Integrated Sample Normalization by MAFFIN Algorithm

High-quality feature selection

Description

Select high-quality features for quantitative analysis.

Usage

FeatureSelection(
  FeatureTable,
  BlankFilter = 2,
  RtRange = c(0, 100),
  QCRSD = 0.25,
  SQCcor = 0.9,
  IntThreshold = 0,
  SampleInCol = TRUE,
  output = FALSE
)

Arguments

FeatureTable	Data frame with features in row and samples in column (default).
BlankFilter	A numeric value. High-quality when mean(sample intensities) > mean(blank intensities) * BlankFilter
RtRange	A numeric vector indicating the range of the defined retention time window, in minute.
QCRSD	A numeric value indicating the relative standard deviation threshold for QC samples.
SQCcor	A numeric value indicating the Pearson's correlation threshold for serial QC samples (recommend: 0.8-0.9).
IntThreshold	A numeric value indicating the feature intensity threshold. Feature is detected when its intensity larger than this value.
SampleInCol	TRUE if samples are in column. FALSE if samples are in row.
output	TRUE will output the result table in current working directory

Details

FeatureTable contains measured signal intensities of metabolic features, with features in row and samples in column (default). The column names should be sample names, and the first row should be sample group names (e.g. control, case).
The first column should be unique feature identifiers. For group names, please use:
"RT" for retention time column;
"QC" for quality control samples between real samples (normal QC samples);
"blank" for blank samples;
"SQC_###" for serial QC samples with a certain loading amount. For example, SQC_1.0 means a serial QC sample with injection volume of 1.0 uL.
An example of FeatureTable is provided as TestingData in this package.

Value

This function will return the original data frame with an extra column named "Quality" to indicate the feature quality.

References

Yu, Huaxu, and Tao Huan. "MAFFIN: Metabolomics Sample Normalization Using Maximal Density Fold Change with High-Quality Metabolic Features and Corrected Signal Intensities." bioRxiv (2021).

Examples

selectedTable = FeatureSelection(TestingData)

Waddlessss/MAFFIN documentation built on Aug. 5, 2023, 8:10 p.m.