Estimates Rocke-Durbin error model parameters for a gene expression dataset.
Usage
fitRockeDurbin(x,theta)
Arguments
x
Numeric data input array to be transformed into universal gene expression 'state' scale. Probes/genes should be in rows, samples/conditions in columns.
theta
Numeric value specifying Normal distribution quantile for identifying groups samples similar enough to be treated as replicates
Value
alpha
background measurement
sd_epsilon
standard deviation of intensity-dependent errors
sd_eta
standard deviation of intensity-independent errors
Details
Estimates error model parameters for Rocke-Durbin model of microarray measurement errors. The model comprises three parameters to be estimated: the background measurement level, an intensity-dependent error term and an intensity-independent error term. This function finds groups of suitably-similar samples from the dataset to be treated as replicates, and estimates model parameters from constitutively-low and constitutively-high measurements.