DSS (version 2.12.0)

dispersion: Acessor functions for the 'dispersion' slot in a SeqCountData object.

Description

Dispersion parameter for a gene represents its coefficient of variation of expressions. It characterizes the biological variations.

Usage

"dispersion"(object)
"dispersion"(object) <- value

Arguments

object

A SeqCountData object.

value

A numeric vector with the same length as number of genes.

Details

If the counts from biological replicates are modeled as negative binomial distribution, the variance (v) and mean (m) should hold following relationship: v=m+m^2*phi, where phi is the dispersion. Another interpretation is that phi represents the biological variations among replicates when underlying expressions are modeled as a Gamma distribution.

See Also

normalizationFactor

Examples

data(seqData)
## obtain
seqData=estNormFactors(seqData, "quantile")
seqData=estDispersion(seqData)
dispersion(seqData)

## assign
dispersion(seqData)=rep(0.1, nrow(exprs(seqData)))