Cancer Genomics Shared Resource
After each array has been processed and scanned, an expression report is generated. The following controls apply to samples submitted with at least 2ug of total RNA undergoing Full chip hybridization from the Core Facility. If the arrays do not meet the standardized set of requirements, the array will be replaced by the Facility with no extra charge to the Principal Investigator.
Requirements for Expression Report (.rpt) Files
Hybridization controls: bioB, bioC, bioD, and cre
These controls are added to the hybridization cocktail, independent of RNA sample preparation and are used to evaluate hybridization efficiency. BioC, bioD, and cre should always be given a present call.
Internal Controls (House Keeping Genes): β-Actin and GAPDH
Β-Actin and GAPDH are used to assess the quality of the synthesis of labeled cRNA. The ratio between the 3’ and 5’ prime signals should be no more than 4, with ideal values being between 1 and 2. Sometimes one of the controls is affected by an artifact. As long as one of the controls is normal, then the sample is good.
Average Background and Intensity
Noise (Q) values are a measure of pixel-to-pixel variation of probe cells on a GeneChip Array; the two main factors contributing to noise are electrical noise of the scanner and sample quality. Background noise will vary among scanners, but should remain consistent among samples analyzed with the same scanner. Background noise values should remain lower then 150 in all cases.
Percent Present (%P) values depend on a multitude of factors including cell/tissue type, biological or environmental stimuli, probe array type, and overall quality of RNA. Percent Present should not vary more than 10-15% within a sample cluster given that those samples being compared are within the same cell or tissue line and subject to similar biological/environmental stimuli.
Scaling and Normalization Factors
Unless requested otherwise, all scaling/normalization at the Microarray Core Facility will undergo the global method of scaling to a target signal of 500, as recommended by Affymetrix. The scale factor (SF) within a sample cluster (under the same conditions as stated for %P requirements) should remain within a three-fold change or lower.