Filter a dataset of features considering pooled quality control samples.
QCs_process.RdGiven a dataframe with feature intensities from a non-targeted experiment and another dataframe containing information about sample types, it filters the dataframe of feature intensities considering some defined cut-off in the quality control samples.
Usage
QCs_process(
featmatrix,
sampletype,
sep_QC = FALSE,
QC_to_merge = NULL,
step1 = TRUE,
step1_cutoff = 50,
step2 = TRUE,
step2_cutoff = 50,
step3 = TRUE,
step3_cutoff = 50,
step4 = FALSE,
step4_cutoff = c(20, 80),
rtrn_filtered_table = TRUE,
remove_results = FALSE,
remove_QC_and_blanks = FALSE
)Arguments
- featmatrix
a dataframe of feature intensities. The first column should contain the feature name, each other columns should be related to an analysis of a sample and contains the feature intensities.
- sampletype
a dataframe containing two or three columns: the first one contains the name of samples reported in the featmatrix. The second column contains, for each sample, one of the following: "REMOVE", for samples to be discharged; "blank", for blank samples; "QC", for pooled quality control samples; "QC_half", for pooled quality control samples diluted by a half; "unknown", for other samples. The third column should contain the sample category that will be considered if sep_QC is TRUE.
- sep_QC
logical. If TRUE, separate QC filtration will be performed. For more details check: Frigerio et al. 2022 https://doi.org/10.3390/molecules27082580.
- QC_to_merge
charter vector. If sep_QC is TRUE, it should contain the names of all the groups of pooled QC samples to consider for the final merge of the filtered features
- step1
logical. Whether to discharge features above a certain relative standard deviation percentage (RSD%) of QCs, defined by step1_cutoff
- step1_cutoff
numeric of length 1. If step1 is TRUE, features with a RSD% of QCs above this value will be considered not suitable.
- step2
logical. Whether to discharge features below a certain QCs count%, defined by step2_cutoff.
- step2_cutoff
numeric of length 1. If step2 is TRUE, features with valid values (not NA) in less than a percentage of QCs specified in this value will be considered not suitable.
- step3
logical. Whether to discharge features above a certain blank contribution%, defined by step3_cutoff.
- step3_cutoff
numeric of length 1. If step3 is TRUE, features with a blank contribution % (calculated as mean blank/mean QCs*100) higher that this value, will be considered not suitable.
- step4
logical. Whether to discharge features not included in a certain range comparing QCs with the QC_half, defined by step4_cutoff.
- step4_cutoff
numeric of length 2. If step4 is TRUE,features that in the QC_half are not within this percentage range comapred to QCs will be considered not suitable.
- rtrn_filtered_table
logical. if TRUE returns the filtered table, if FALSE returns the non-filtered table (useful only if you want to check the parameters such as QC CV% or blank contribution in the features that would be filtered)
- remove_results
logical. If TRUE, returns the table of features, if FALSE, the last columns will report the results such as QC CV% and blank contribution
- remove_QC_and_blanks
logical. At the end of the processing, if TRUE it will remove columns containing QCs and blanks (so, if also remove_results is TRUE and rtrn_filtered_table is TRUE, at the end you will have a great table of feature intensities with only samples ready for the next processing)