evaluateK: Evaluate normalization using K slope groups
In rhondabacher/SCnorm: Normalization of single cell RNA-seq data

evaluateK

R Documentation

Evaluate normalization using K slope groups

Description

Median quantile regression is fit for each gene using the normalized gene expression values. A slope near zero indicate the sequencing depth effect has been successfully removed. Genes are divided into ten equally sized groups based on their non-zero median expression. Slope densities are plot for each group and estimated modes are calculated. If any of the ten group modes is larger than .1, the K is not sufficient to normalize the data.

Usage

evaluateK(
  Data,
  SeqDepth,
  OrigData,
  Slopes,
  Name,
  Tau,
  PrintProgressPlots,
  ditherCounts
)

Arguments

`Data`	matrix of normalized expression counts. Rows are genes and columns are samples.
`SeqDepth`	vector of sequencing depths estimated as columns sums of un-normalized expression matrix.
`OrigData`	matrix of un-normalized expression counts. Rows are genes and columns are samples.
`Slopes`	vector of slopes estimated in the GetSlopes() function. Only used here to obtain the names of genes considered in the normalization.
`Name`	plot title
`Tau`	value of quantile for the quantile regression used to estimate gene-specific slopes (default is median, Tau = .5 ).
`PrintProgressPlots`	whether to automatically produce plot as SCnorm determines the optimal number of groups (default is FALSE, highly suggest using TRUE). Plots will be printed to the current device.
`ditherCounts`	whether to dither/jitter the counts, may be used for data with many ties, default is FALSE.