techVarSub: Technical noise analysis for single cell RNAseq
In wjawaid/bglab: Berthold Gottgens lab scRNA seq

Description Usage Arguments Details Value Author(s)

Normalises and performs technical noise analysis for single cell RNAseq(Brennecke et al.)

1
2
3

techVarSub(object, data.ercc = NULL, cvThresh = 0.3, quant = 0.8,
  minBiolDisp = 0.5^2, fdr = 0.1, data.lengths = NULL,
  ercc.lengths = NULL, meanForFit = NULL)

`object`	SCD object
`data.ercc`	A matrix of counts with cells in columns and ERCC spike-ins in rows. The columns must mutch those in the gene count table passed in "data".
`cvThresh`	Defalt 0.3. The CV threshold to use for choosing cells to fit. See quant.
`quant`	Default 0.8. The quantile of data to fit so that linear model is not fitted to plateau at lower gene expression values. See Breenecke et al. for further details
`minBiolDisp`	Default 0.25. Assumed biological dispersion. See paper.
`fdr`	Deafult .1. False Discovery Rate for chi-squared test.
`data.lengths`	Gene lengths in kilobases for PK "per kilobase normalisation"
`ercc.lengths`	ERCC lengths in kilobases for PK "per kilobase normalisation"
`meanForFit`	Provide a minimum mean for fit. (OPTIONAL)

Normalises and performs a technical noise analysis for single cell RNAseq. Give raw counts. Data is normalised using size-factor normalisation identical to DESeq2, then a technical noise analysis is perfromed as described in brennecke et al.

Returns a list

Wajid Jawaid

wjawaid/bglab documentation built on May 4, 2019, 6:33 a.m.