pvca: A function for principal variance component analysis

In wangjiaxuan666/fixbatch: The goal of packages is fixing the batch effect in data

A function for principal variance component analysis

Description

The function is written based on the 'pvcaBatchAssess' function of the PVCA R package(http://watson.nci.nih.gov/bioc_mirror/packages/release/bioc/manuals/pvca/man/pvca.pdf). and adjust by Donghyung Lee for changed slightly to make it more efficient and flexible for sequencing read counts data(https://github.com/dleelab/pvca). But it didn't update since 4 years ago. My new R packages fixbatch need the function to estimate factor's partition of the total variability. So I fork the packages and Adjust in my own style.

Usage

pvca(counts, meta, threshold = 0.7, inter = TRUE)

Arguments

counts	The Normalized(e.g. TMM)/ log-transformed reads count matrix from sequencing data (row:gene/feature, col:sample)
meta	The Meta data matrix containing predictor variables (row:sample, col:predictor)
threshold	The proportion of the variation in read counts explained by top k PCs. This value determines the number of top PCs to be used in pvca.
inter	TRUE/FALSE - include/do not include pairwise interactions of predictors

Value

std.prop.val The vector of proportions of variation explained by each predictor.

wangjiaxuan666/fixbatch documentation built on Jan. 25, 2024, 4:39 p.m.