Sample and Chip Quality Assessment

Agilent BioAnalyzer

The Agilent BioAnalyzer is used to assess the quality of the total RNA sample.  The BioAnalyzer performs automated electrophoretic separations similar to those obtained by running a gel electrophoresis.  The results are inspected to ensure that both ribosomal peaks are intact and that no degradation has occurred.  Normally we do not conduct gene chip experiments with poor quality total RNA samples.

Spectrophotometry

            260/280 absorbance readings are measured for both total RNA and biotinylated cRNA. Acceptable 260/280 ratios fall in the range of 1.8 to 2.1. Ratios below 1.8 indicate possible protein contamination. Ratios above 2.1 indicate presence of degraded RNA, truncated cRNA transcripts, and/or excess free nucleotides (from Affymetrix Data Analysis Fundamentals Manual).  

Microarray Suite

            Before any analysis is performed on a scanned chip, the image is visually examined to check for scratches or other artifacts.  If such artifacts exist and are relatively small, they can be masked out so that the affected probe pairs are not included in the analysis.  However, if the artifacts are too large or too numerous, the expression chip is rerun at no additional cost to the client.

            The Microarray Suite expression analysis algorithms, which calculate the signal intensity values and detection calls, also calculate several parameters relating to the quality of the expression chip, background intensity, noise and scale factor being the most important.  The Background intensity, derived from the intensity values of the lowest 2% of cells on the chip, establishes an overall baseline intensity to be subtracted from all cells before gene expression levels are calculated.  Noise is derived from the standard deviation of the Background intensity measurement.  Affymetrix suggests that, for best analytical results, Background intensities and Noise values should not exceed 100 and 4, respectively.  The scale factor is inversely related to the overall brightness of the chip.  A more detailed mathematical explanation of these parameters can be provided upon request.

Another indication of whether a chip contains excessive levels of background and noise involves spike control detection calls.  The BioB control cRNA is spiked in at the detection threshold (1.5 pM) and should receive ‘present’ detection call in approximately 50 percent of all samples; however, too much Noise in a sample may prevent low abundance transcripts, such as the BioB spike control, from being detected.

The Microarray Suite expression report also contains information relating to chip quality.  Most important are the 3’ end to 5’ end signal intensity ratios for the housekeeping genes.  Chips include probe sets derived from the 3’ end and from the 5’ end of all control transcripts.  Since degradation is more likely to occur at the 5’ end of the mRNA molecule, which is not protected by a polyA tail, a high ratio may indicate sample degradation.  Affymetrix states that housekeeping gene 3’ to 5’ ratios should ideally be no higher than 4.  The expression report also gives the percentage of the genes on the chip with a detection call of ‘present’.  For most expression chips, this percentage should be between 40 and 60.  Low percent present may also indicate degradation or incomplete synthesis.