xMSwrapper.XCMS.matchedFilter: xMSwrapper.XCMS.matchedFilter

In yufree/xMSanalyzer: xMSanalyzer: R package for data analysis, comparison, and annotation of metabolomics data.

Description

Wrapper function for XCMS based on matched filter alignment algorithm, evaluate.Features,evaluate.Samples,merge.Results, and feat.batch.annotation

Usage

xMSwrapper.XCMS.matchedFilter(cdfloc, XCMS.outloc, xMSanalyzer.outloc, 
    step.list = c(0.001, 0.01, 0.1), mz.diff.list = c(0.001, 0.01, 
        0.1), sn.thresh.list = c(10), max = 50, bw = c(10), minfrac.val = 0.5, 
    minsamp = 2, mzwid = 0.25, sleep = 0, retcor.family = "symmetric", 
    retcor.plottype = "mdevden", groupval.method = "medret", numnodes = 2, 
    run.order.file = NA, max.mz.diff = 15, max.rt.diff = 300, merge.eval.pvalue = 0.2, 
    mergecorthresh = 0.7, deltamzminmax.tol = 10, num_replicates = 2, 
    subs = NA, mz.tolerance.dbmatch = 10, adduct.list = c("M+H"), 
    samp.filt.thresh = 0.7, feat.filt.thresh = 50, cormethod = "pearson", 
    mult.test.cor = TRUE, missingvalue = 0, ignore.missing = TRUE, 
    sample_info_file = NA, refMZ = NA, refMZ.mz.diff = 10, refMZ.time.diff = NA, 
    void.vol.timethresh = 30, replacezeroswithNA = TRUE, scoreweight = 30, 
    filepattern = ".cdf", charge_type = "pos", minexp.pct = 0.1, syssleep = 0.5)

Arguments

cdfloc	The folder where all CDF/mzXML files to be processed are located. eg: "C:/CDF/". Use "cdfloc=NA" if the XCMS results already exist in XCMS.outloc.
XCMS.outloc	The folder where XCMS output will be written. eg: "C:/XCMSoutput/"
xMSanalyzer.outloc	The folder where xMSanalyzer output will be written. eg: "C:/xMSanalyzeroutput/"
step.list	list containing values for the step size
mz.diff.list	list containing values for the minimum difference for features with retention time overlap
sn.thresh.list	list containing values for signal to noise ratio cutoff variable
max	Value for maxnimum number of peaks per EIC variable
bw	bandwidth value list. eg: c(10,30)
minfrac.val	minimum fraction of samples necessary in at least the sample groups for it to be a valid group
minsamp	minimum number of samples necessary in at least one of the sample groups for it to be a valid group
mzwid	width of overlapping m/z slices to use for creating peak density chromatograms and grouping peaks across samples
sleep	seconds to pause between plotting successive steps of the peak grouping algorithm. Peaks are plotted as points showing relative intensity. identified groups are flanked by dotted vertical lines.
retcor.family	Used by matchedFilter alignment method. Use "gaussian" to perform fitting by least squares without outlier removal. Or "symmetric" to use a redescending M estimator with Tukey's biweight function that allows outlier removal.
retcor.plottype	Used by both matchedFilter and centWave alignment methods. eg: "deviation" or "mdevden"
groupval.method	Conflict resolution method while calculating peak values for each group. eg: "medret" or "maxint"
numnodes	Number of computing nodes to use. eg: 1
run.order.file	Name of a tab-delimited file that includes sample names sorted by the order in which they were run (sample names must match the CDF file names)
max.mz.diff	+/- mz tolerance in ppm for feature matching
max.rt.diff	retention time tolerance for feature matching
merge.eval.pvalue	Threshold for defining signifcance level of the paired t-test or the Pearson correlation during the merge stage in xMSanalyzer. The p-value is used to determine whether two features with same m/z and retention time have identical intensity profiles.
mergecorthresh	Correlation threshold to be used during the merge stage in xMSanalyzer to determine whether two features with same m/z and retention time have identical intensity profiles.
num_replicates	Number of replicates per sample
subs	If not all the CDF files in the folder are to be processed, the user can define a subset using this parameter.
mz.tolerance.dbmatch	m/z threshold for database matching
adduct.list	List of adducts for matching m/zs in KEGG. eg: c("M+H","M+H-H2O")
samp.filt.thresh	Threshold for filtering samples based on Pearson correlation between technical replicates. eg: 0.7
feat.filt.thresh	Threshold for filtering samples based on percent intensity difference or coefficient of variation. eg: 50
cormethod	Method for determing correlation between technical replicates. eg: "pearson" or "spearman
mult.test.cor	Should Bonferroni multiple testing correction method be applied for comparing intensities of overlapping m/z? Default: FALSE
missingvalue	How are missing values represented? eg: 0 or NA
ignore.missing	Should the missing values be ignored while computing pearson correlation. eg: TRUE or FALSE
deltamzminmax.tol	Maximum allowed delta ppm between mz.min and mz.max. Eg: 10
refMZ	Full path of the file with m/z of the targeted chemicals to search for. If the value is "NA", then the list of metabolites in the data(example_target_list) is used. The input file should be formatted as data(example_target_list): Column A: m/z of the targeted metabolite (required) Column B: retention time (Optional) Column C: Name of the metabolite (required)
refMZ.mz.diff	The ppm tolerance to search for targeted metabolites/features in xMSanalyzer. Default: 10
refMZ.time.diff	The time tolerance to search for targeted metabolites/features in xMSanalyzer. Default: NA
void.vol.timethresh	Time threshold for void volume. The program searches for the void volume cutoff within the defined time limit. Default: 30
replacezeroswithNA	Should 0s be treated as missing values during ComBat (TRUE or FALSE). Default: TRUE
scoreweight	The w parameter in the scoring function defined in the xMSanalyzer manuscript. Uppal 2013, BMC Bioinformatiocs. Default: 30
sample_info_file	File listing the order in which the samples were run. The format of the file should be as follows: Column A:File names matching the CDF/mzXML files Column B: Sample ID Column C: Batch number (This column should be labeled "Batch") Column D: Additional covariates to adjust for (Optional)
filepattern	File format of spectral data files. eg: ".cdf", ".mzXML"
charge_type	Ionization mode. "pos" or "neg" Default: "pos"
minexp.pct	Minimum fraction of samples in which the signal should be present. Default: "0.1"
syssleep	Sleep time in between KEGG REST queries. Default: "0.5"

Details

The wrapper function includes five stages to utilize information from the technical replicates to optimize the data extraction process, enhance data quality, search for targeted features/metabollites, perform QA & QC evaluations including batch-effect evaluation & correction. : 1) features are extracted using different parameters 2) results from each parameter setting are evaluated for sample quality & feature consistency, 3) filtered results from individual settings are merged to obtain a combined feature table; the optimization score that takes into account the number of features and average CV (see Uppal 2013) is used to determine the most optimal set of parameters. 4) A targeted feature table using the list of m/z in the "refMZ" file is generated 5) Quality measures of each featue includes: number of samples including replicates with non-missing values (NumPres.All.Samples), number of biological samples for which more than 60 median coefficient of variation (CV) within technical replicates summarized across all samples; Qscore (Quality score which is calculated using NumPres.All.Samples, NumPres.Biol.Samples, median CV, and delta ppm between mz.min and mz.max; Higher is better) Users have the option to filter poor quality samples and features based on correlation between technical replicates and feature reproducibility measures such as PID or CV, respectively.

Value

A list is returned.

XCMS.merged.res	Merged feature table, P1 U P2 where P1 and P2 are two sets of parameter settings. The "QRscore" is defined as the ratio of percentage of biological samples for which more than 50 percent of technical replicates have a signal and median PID or CV. QRscore=Percent good samples/median PID or CV
XCMS.ind.res	List with results from individual parameter settings.
XCMS.ind.res.filtered	List with results from individual parameter settings after filtering.
annot.res	List with annotation results after merging results.
feat.eval.ind	List with feature evaluation results based on CV or PID at each parameter setting.
sample.eval.ind	List with sample evaluation results at each parameter setting based on correlation between technical replicates.

Outputs XCMS results at each parameter settings to XCMS.outloc and the following to xMSanalyzer.outloc: -feature consistency results -sample evaluation results -feature list after merging results from parameter settings A and B -merge summary

Author(s)

Karan Uppal <[e-mail]>

yufree/xMSanalyzer documentation built on May 4, 2019, 6:35 p.m.