Reference gene names not used in calibration_factors

[iimog]

Thanks for keeping easyqpcr alive. I recently stumbled over the issue, that even though I was passing the names of reference_genes to calibration_factors, other genes were actually used as reference. I tracked it down to the internal call to normalize_data where there is the hard-coded assumption, that the first columns contain the reference genes:

easyqpcr2/R/calibration_factors.R

Lines 127 to 129 in 4810ba6

	# Set the column position(s) of the reference gene(s),
	# assuming they are the first genes in the data frame
	, Refposcol = 1:length(reference_genes)

This assumption is not communicated in the documentation

easyqpcr2/R/calibration_factors.R

Lines 15 to 16 in 4810ba6

	#' @param reference_genes
	#' Character vector with the name(s) of the reference gene(s)

The problem is, that this implementation does not produce errors, but rather works in an unexpected way.

I added this line to calibration_factors to actually get the reference genes positions and pass them to normalize_data:

# positions are shifted by -1 because the first column is dropped later
reference_genes_pos <- which(colnames(tmp_df) %in% reference_genes)-1

[dsrodriguezl]

Thanks for flagging this up, I probably just overlooked that when adapting the code of the original easyqpcr package into the new function. I will take a look at your suggestion and adapt it if I do not see any problem.