Quality Control

We routinely test all probes on Genechip test arrays. The following 5 quality control measures contribute to our acceptance/rejection of microarray data:

1. Hybridization of the B2 oligonucleotide.
   The border of the microarray chip should be bright and show homogeneous staining.
   


2. Average Difference (AvgDiff) values of 5’ and 3’ probe sets corresponding to murine GAPDH (M32599) and beta-actin (M12481).
   The AvgDiff derived from the probe sets corresponding to the 5’ and 3’ ends of each gene. We accept chips when the 5’probe set AvgDiff is at least half of the value of the 3’probe set AvgDiff.
   


3. AFFX-BioB-3_at, AFFX-BioC-5_at, and AFFX-CreX-5_at spiked in controls:
   AvgDiff values of 5’ and 3’ probe sets should follow rule 2, AND show an increase in quantity from BioB to CreX, as this demonstrates the sensitivity of hybridization.
   


4. RawQ values as a measure of noise.
   RawQ values result from small variations in the digitized signal as the scanner samples the surface of each probe array. The level of noise is calculated by the software, and then used as a criteria to determine the significance of differences between Perfect Match (PM) and Mismatch (MM) probe cells. Within a given set of experiments, RawQ values should be within the range of 5.0 points of each other.
   


5. The scaling factor SF that is used to normalize the signal across the entire chip to an arbitrary Target Intensity.
   The Scaling Factor (SF) for each given experiment should be within a 2-3 fold range.
   

We will accept chip data that pass criteria 2 and 3, and at least two of the other three criteria. In addition, we perform a visual inspection of the chip for scratches, cracks, and presence of any contaminants on the chip surface. For minor defects, we manually block defective probe sets, or repeat the hybridization using a new chip.

Back to top^

Page last modified: 20-May-2003