Table 3.

The table summarizes qRT-PCR results (mean raw threshold cycle [Ct]-values) of four gene targets (CDKN1A, FDXR, DDB2 and GAPDH) for the saliva samples.

Genes	Pre-amplification	Measurements				Interpretation
		# Samples/comparisons	Mean Ct values		Difference observed vs expected	Manufacturer predicts success	Our findings: works?	Agreement?
			Observed	Expected
CDKN1A	0 ×	12	33.1			Yes	Yes	Yes
	10 ×	12/12	23.2	23.1	0.09
	14 ×	12/12	19.3	19.1	0.24
DDB2	0 ×	12	36.1			No	No
	10 ×	12/12	27.9	26.1	1.8
	14 ×	12/12	24.5	22.1	2.33
FDXR	0 ×	12	37.4			No	Yes	No
	10 ×	12/12	28.4	27.4	1.02
	14 ×	12/5	23.9	23.4	0.49
GAPDH	0 ×	12	33.5			Yes	No
	10 ×	12/12	29.3	23.5	5.78
	14 ×	12/12	26.2	19.5	6.69
In total		12/89

Number of samples and number of comparisons per gene and pre-amplification pattern are depicted in italics. Shown are the measured/observed mean Ct values for unamplified samples as well as for samples after 10 × and 14 × pre-amplification. The column for expected mean Ct values shows the calculated Ct values when estimating a 100% linear and unbiased 10 × and 14 × pre-amplification (steps by 10 and 14, taking the mean Ct value without pre-amplification). The ΔCt _{(observed vs expected)} is shown for each gene and each pre-amplification. We categorized the data by whether unbiased pre-amplification should have worked according to the manufacturer (Ct value without pre-amplification < 35, yes) or not (Ct value without pre-amplification > 35, no). The data are further presented if pre-amplification worked unbiased (yes) or not (no) according to the amplification biases through deviating expression patterns between amplified and un-amplified material.