LoadOntoPCs: Calculate Test Data PCs from Training-Data Estimated Loadings

In gabrielodom/pathwayPCA: Integrative Pathway Analysis with Modern PCA Methodology and Gene Selection

Calculate Test Data PCs from Training-Data Estimated Loadings

Description

Given a list of loading vectors from a training data set, calculate the PCs of the test data set.

Usage

LoadOntoPCs(design_df, loadings_ls, sampleID = c("firstCol", "rowNames"))

Arguments

design_df	A test data frame with rows as samples and named features as columns
loadings_ls	A list of p \times d loading vectors or matrices as returned by either the SuperPCA_pVals, AESPCA_pVals, or ExtractAESPCs functions. These lists of loadings will have feature names as their row names. Such feature names must match a subset of the column names of design_df exactly, as pathway-specific test-data subsetting is performed by column name.
sampleID	Are the sample IDs in the first column of design_df or in accessible by rownames(design_df)? Defaults to the first column. If your data does not have sample IDs for some reason, set this to rowNames.

Details

This function takes in a list of loadings and a training-centered test data set, applies over the list of loadings, subsets the columns of the test data by the row names of the loading vectors, right-multiplies the test-data subset matrix by the loading vector / matrix, and returns a data frame of the test-data PCs for each loading vector.

Value

A data frame with the PCs from each pathway concatenated by column. If you have the tidyverse loaded, this object will display as a tibble.

Examples

  ###  Load the Data  ###
  data("colonSurv_df")
  data("colon_pathwayCollection")

  ###  Create -Omics Container  ###
  colon_Omics <- CreateOmics(
    assayData_df = colonSurv_df[, -(2:3)],
    pathwayCollection_ls = colon_pathwayCollection,
    response = colonSurv_df[, 1:3],
    respType = "survival"
  )

  ###  Extract AESPCs  ###
  colonSurv_aespc <- AESPCA_pVals(
    object = colon_Omics,
    numReps = 0,
    parallel = TRUE,
    numCores = 2,
    adjustpValues = TRUE,
    adjustment = c("Hoch", "SidakSD")
  )

  ###  Project Data onto Pathway First PCs  ###
  LoadOntoPCs(
    design_df = colonSurv_df,
    loadings_ls = colonSurv_aespc$loadings_ls
  )

gabrielodom/pathwayPCA documentation built on July 10, 2023, 3:32 a.m.