tissue: biopsy taken from occult metastasis gender: male
Extracted molecule
total RNA
Extraction protocol
Not provided
Label
Cyanine-3-labeled cRNA
Label protocol
100 ng total RNA per sample were introduced into a reverse transcription with subsequent in vitro transcription (RT-IVT) reaction. Prior to RT-IVT, the total RNA samples were spiked with in vitro synthesized polyadenylated transcripts (One-Color RNA Spike-In Mix, Agilent Technologies) which serve as an internal labeling control for linearity, sensitivity and accuracy. The spiked total RNA was reverse transcribed into cDNA and then converted into Cyanine-3 labeled cRNA (Low Input Quick-Amp Labeling Kit One-Color, Agilent Technologies).
Hybridization protocol
Following cRNA clean-up and quantification 600 ng of each Cyanine-3-labeled cRNA sample was fragmented and prepared for one-color-based hybridization (Gene Expression Hybridization Kit, Agilent Technologies). Labeled cRNA samples were hybridized at 65°C for 17 hrs on separate Agilent SurePrint G3 Human Gene Expression 8x60K v2 Microarrays (AMADID 039494). Afterwards, microarrays were washed with increasing stringency using Gene Expression Wash Buffers (Agilent Technologies) followed by drying with acetonitrile (SIGMA).
Scan protocol
Fluorescent signal intensities were detected with Scan Control A.8.4.1 software (Agilent Technologies) on the Agilent DNA Microarray Scanner and extracted from the images using Feature Extraction 10.7.3.1 software (Agilent Technologies).
Description
RS-315_0006
Data processing
The quantile normalized, log2 transformed data were averaged across replicates, resulting in the columns group # (normalized). A pairwise comparison was analyzed. Group lymphogen was used as reference group and RNA expression of group occult was compared to this baseline. To filter for genes with a robust differential expression between two compared groups, several parameters are taken into account: Robust detection: A probe is only considered if it is reliably detectable (“Detected” call, flag information from the Feature Extraction software) in all samples of one of the two compared groups. Specifically, if filtering for induction in group x compared to group y, the probe must be called “Detected” in all samples of group x. If filtering for repression, it must be called “Detected” in all samples of the reference group y. Statistical significance: Two filtering approaches were used that apply different stringency levels to the data set: a) Stringent filtering: A probe is only classified as induced in a specific comparison if its corrected p-value is ≤ 0.05 and if it has a Fold Change value ≥ 1.5 and 2, respectively. Analogously, a probe is classified as repressed if its corrected p-value is ≤ 0.05 and its Fold Change is ≤ -1.5 and -2, respectively. By using this stringent filtering, the multiple testing error is taken into account, and the false discovery (FDR) rate is controlled at a maximum level of 5%. b) Non-stringent filtering: Stringency was reduced by using the non-corrected instead of the corrected p-value. Using this approach, a probe is only classified as induced in a specific comparison if its (non-corrected) p-value is ≤ 0.05, and if it has a Fold Change value ≥ 1.5 and 2, respectively. Analogously, a probe is classified as repressed if its (non-corrected) pvalue is ≤ 0.05 and its Fold Change is ≤ -1.5 and -2, respectively. By using this non-stringent filtering, the multiple testing error is not taken into account. Finally, several analysis tools from the PANTHER website, release 8.1 (accessible through http://www.pantherdb.org/) were applied for functional evaluation of differential regulation.
