Arranges a vector of intensities into a form amenable to analysis by the DDHF and also can restrict the number of genes analysed and also obtain a random sample
cDNAdata(data.vect,cdnalength,datasize,ng,nrep)
The data vector in the proper format to perform Data-Driven Haar-Fisz algorithm
The data vector of intensities
Only considers the first cdnalength genes for analysis
Needs to be a power of two. The number of genes that get
randomly sampled from the first cdnalengthq and then
subsequently transformed.
The total number of genes described by data.vect
The number of replicates (should be a power of two) contained
in data.vect
Efthimios Motakis <e.motakis@bris.ac.uk>
The $$J = ng \times nrep$$ vector data.vect should contain first the
intensities of the first replicate of the ng genes',
then the second replicate of all of the ng genes in the
same order, and so on.
This function first puts the one dimensional data.vect into
a matrix form with ng rows and nrep columns (so the
row number indices the gene and the column number the replicate).
Then the first cdnalength rows are extracted and kept, the remaining
rows are discarded.
Then datasize rows at random are extracted and kept and
the remaining rows are discarded.