In jennyjyounglee/AclustsCCA: A Cluster Based Sparse Canonical Correlation Analayis to Test Associations Between A Multi-Pollutant Mixture and High-Dimentional DNA Methylation

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

1. What is Partial Residuals?

Partial residuals are .... [@sofer2012multivariate]

2. Input Parameters for Partial Residuals

| data A data matrix with \eqn{n} rows.

| Z A \eqn{n} by \eqn{r} confounder data matrix, where \eqn{n} is sample size and \eqn{r} is number of potential confounders.

3. Obtain partial residuals to adjust for potential confounders

When implementing SparseCCA, partial residuals are used to adjust for potential confounders. Again, computing partial residuals have to be ran only once. I personally suggest to run this part separately and save it for future use.

# Load sample data
data(sample.data)
DATA.X <- sample.data$DATA.X # row: subjects (n), column: exposures (p)
DATA.Y <- sample.data$DATA.Y # row: subjects (n), column: CpG sites (q)
DATA.Z <- sample.data$DATA.Z # row: subjects (n), column: confounders (r)

X.resid <- partial.residual(data=DATA.X,Z=DATA.Z,nthread=1)
Y.resid <- partial.residual(data=DATA.Y,Z=DATA.Z,nthread=1)
save(X.resid,Y.resid,"partial.residual.RData")

References

jennyjyounglee/AclustsCCA documentation built on June 15, 2022, 7:45 p.m.