acPCA: Perform AC-PCA for simultaneous dimension reduction and...

In YuWang28/acPCoA: Adjustment for Confounding Factors using Principal Coordinate Analysis

Description

Perform AC-PCA for simultaneous dimension reduction and adjustment for confounding variation

Usage

acPCA(
  X,
  Y,
  nPC,
  lambda,
  eval = F,
  numPerm = 100,
  alpha = 0.05,
  plot = T,
  centerX = T,
  centerY = T,
  scaleX = F,
  scaleY = F,
  kernel = c("linear", "gaussian"),
  bandwidth = 1
)

Arguments

X	the n by p data matrix, where n is the number of samples, p is the number of variables. Missing values in X should be labeled as NA. If a whole sample in X is missing, it should be removed.
Y	the n by q confounder matrix, where n is the number of samples, q is the number of confounding factors. Missing values in Y should be labeled as NA.
nPC	number of principal components to compute
lambda	the tuning parameter, non-negative.
eval	True or False. eval=T evaluates the significance of the PCs. Default is F.
numPerm	the number of permutations to evaluate the significance of the PCs. Default is 100.
alpha	the significance level. Default is 0.05. If the kth PC is not significant, we don't consider the PCs after it. If the eigenvalue and variance explained by the PCs give inconsistent results, we choose the maximum number of significant PCs.
plot	True or False. plot=T generates the plots. Default is True.
centerX	center the columns in X. Default is True.
centerY	center the columns in Y. Default is True.
scaleX	scale the columns in X to unit standard deviation. Default is False.
scaleY	scale the columns in Y to unit standard deviation. Default is False.
kernel	the kernel to use: "linear", "gaussian".
bandwidth	bandwidth h for Gaussian kernel. Optional.

Value

The principal components and the projected data

v	the principal components, p by nPC matrix
Xv	the projected data, i.e. X times v
eigenvalueX	eigenvalues for the PCs
varianceX	variance explained by the PCs
varianceX_perc	percentage of total variance in X explained by the PCs. If eval=F, NA is returned.
eigenvalueXperm	eigenvalues for the PCs, permutation. If eval=F, NA is returned.
varianceXperm	variance explained by the PCs, permutation. If eval=F, NA is returned.
sigPC	the significant PCs. If eval=F, NA is returned.
...	Input parameters for the function

YuWang28/acPCoA documentation built on Dec. 18, 2021, 8:20 p.m.