Overall, microarray spot p-value appeared to have a considerable effect on the correlations between array and qPCR results (Fig. 3a).
As seen in Fig. 2, for those eight microarray spots, the value of the background polyA hybridization Bi appears to be much larger than the median value of Hi among all spots, i.e., if the polyA signal is greater than 12000, then the value of Hi will always be greater than ~24000 phosphorimager units.
Only microarrays that showed a scatter plot of spot values from scrapie vs. mock with a slope of approximately one were considered for subsequent analysis.
To evaluate the quality of the data from each scanned microarray, a scatter plot was created by plotting mock verses scrapie-infected normalized and background subtracted spot values.
The expression value is usually the pre-processed intensity value (i.e., the value assigned to the spot as it has been processed by the software of the specific experimental platform used, for example, following background subtraction for a microarray spot).
Microarray spot features were identified, quantified, and assessed using BlueFuse for Microarrays v2 (BlueGnome).
The majority of the microarray spots had an M-value of around 0, which, together with the fact that we compared large datasets with over 9000 measurements, might inflate the p-value of the correlation coefficient and distort a statistical comparison between the generations.
In general, genes exhibiting late Ct values corresponded to low intensity microarray spots: 75.6% of genes with Ct values above the median exhibited below-median spot intensity, and thus likely represent genes with low expression levels.
Then we calculated the mean and SD (standard deviation) for the log2(R1/R2) values of all microarray spots.
Normalized data was log transformed and microarray spots in the t-test combined with ratio values double difference were regarded as differentially expressed genes.
