Normalization: Normalization for DeMix

In DeMix: Deconvolution Models for Mixed Transcriptomes from Heterogeneous Tumor Samples

Description

A normalization function for quantile/total/median normalization. The preparation work – quantile normalization before DeMix (Bayesian deconvolution models for both RNA-seq read counts and microarray expressions).

Usage

DeMix.Normalization(input,
                    design,
                    method=c("total", "quantile", "median"))

Arguments

input	numeric G*S matrix of gene expressions before normalization and selection, where G is the number of genes and S is the number of samples
design	numeric vector of length S specifying whether the sample type is normal or mixed tumor
method	character string specifying the normalization method

Value

Returns same-sized numeric matrix of normalized gene expressions.

Author(s)

Jaeil Ahn, P. Roebuck [e-mail]

References

Ahn J, Yuan Y, Parmigiani G, Suraokar MB, Diao L, Wistuba II, & Wang W.
DeMix: deconvolution for mixed cancer transcriptomes using raw measured data
Bioinformatics (2013) 29 (15): 1865-1871
https://doi.org/10.1093/bioinformatics/btt301

Wu H, Wang C, & Wu Z.
A new shrinkage estimator for dispersion improves differential expression detection in RNA-seq data
Biostatistics (2013) 14 (2): 232-243
https://doi.org/10.1093/biostatistics/kxs033

Examples

data(ntot_NT_Liver)
input <- ntot_NT_Liver.df
design <- c(rep(0, 60), rep(1, 151))   # 60 normal, then 151 tumor samples
input.norm <- DeMix:::DeMix.Normalization(input, design, method="quantile")

DeMix documentation built on May 2, 2019, 5:19 p.m.