RestrictedPCA: RestrictedPCA.

In MetabolomicsBasics: Basic Functions to Investigate Metabolomics Data Matrices

RestrictedPCA.

Description

RestrictedPCA combines an ANOVA based on 'fmod' and restricts a PCA using the ANOVA result as a filter.

Usage

RestrictedPCA(
  dat = NULL,
  sam = NULL,
  use.sam = NULL,
  group.col = NULL,
  text.col = NULL,
  fmod = NULL,
  sign.col = NULL,
  p.adjust.method = "none",
  P = 0.01,
  pcaMethods.scale = "pareto",
  n.metab.min = 20,
  ...
)

Arguments

dat	Metabolite matrix (samples x metabolites).
sam	Sample definition dataframe.
use.sam	Numeric index vector (or logical) to select specific samples to be included in the analysis or NULL to include all.
group.col	Column used for legend creation (column name from sam).
text.col	Column used for text annotation of data points (column name from sam).
fmod	ANOVA model to calculate before PCA.
sign.col	Which column(s) of the ANOVA result shall be used for P-value filtering (specify column names or leave on NULL to filter on all).
p.adjust.method	Method use to adjust P-values (e.g. none, BH or bonferroni).
P	P-value threshold used as a cutoff after P-value adjustment.
pcaMethods.scale	pcaMethods scale parameter (usually pareto for metabolite data).
n.metab.min	Minimum number of metabolites kept for PCA calculation (even if they exceed P).
...	Handed through to PlotMetabolitePCA.

Details

‘fmod' should be something like ’GT*TR+Batch' to perform an ANOVA with these factors defined as columns in sam.

Value

Will generate a PCA plot (generated by PlotMetabolitePCA internally) restricted based on an ANOVA result based on MetaboliteANOVA.

Examples

# load raw data and sample description
raw <- MetabolomicsBasics::raw
sam <- MetabolomicsBasics::sam
# standard behavior
RestrictedPCA(dat = raw, sam = sam, group.col = "GT")
## Not run: 
# apply multiple testing using a strict P-value cutoff,
# dont show a legend but plot group mean values and sd's as overlay
RestrictedPCA(
  dat = raw, sam = sam, group.col = "GT", p.adjust.method = "BH", P = 10^-10,
  fmod = "GT+Batch+Order", sign.col = "GT", medsd = T, legend.x = NULL
)
# limit to a subset of samples, switching the ANOVA selection of by setting P=1
# and adding text (from \code{sam}) to each data point
RestrictedPCA(
  dat = raw, sam = sam, use.sam = which(sam$GT %in% c("Mo17", "B73")), group.col = "GT",
  fmod = "GT+Batch+Order", P = 1, sign.col = "GT", legend.x = NULL, text.col = "Batch"
)

## End(Not run)

MetabolomicsBasics documentation built on May 29, 2024, 9:02 a.m.