Keck-UNM Genomics Resource

Frequently Asked Questions for Affymetrix Microarray Gene Expression Experiments

Prepared by the Keck-UNM Genomics Resource, CRF 118, Tel. 272-5564.

For additional information see our website: http://hsc.unm.edu/som/micro/genomics/

This document provides guidelines for researchers contemplating the use of Affymetrix microarrays to monitor changes in gene expression in experimentally manipulated (e.g. treated vs. untreated) samples. All researchers are urged to consult with the KUGR staff before beginning to prepare or analyze samples. We can assist with experimental design and, if necessary, put you in touch with expert biostatisticians.

Download the FAQ document (PDF File).

Frequently Asked Questions: (click on the link to jump down the page)

How much RNA do I need?
What will the KUGR Facility provide?
How much will it cost?
Do I really need duplicates?
How good is the Affymetrix data?
What controls are used in the analysis?
How will the data be analyzed?
What is the 'day effect'?
Who will help me analyze my microarray data?
How can I verify my microarray results?
What other services does the KUGR facility provide?
What information should I include in my grant application?
What things should I avoid in grant applications containing microarray experiments?

How much RNA do I need?

The optimum starting material is about 1 µg of high quality, total RNA (approx. 2x105 human cells). However, as little as 10 ng of total RNA can be successfully analyzed, but may require additional amplification steps (higher cost).

What will the KUGR Facility provide?

We provide a full service, RNA to Data analysis package that is perfect for new users or researchers with limited experience working with RNA. Researchers provide us with RNA and the appropriate Affymetrix GeneChips (purchased at reduced cost through the CRTC). We will analyze the RNA for quality and amount, perform the cDNA and cRNA synthesis and necessary clean-up steps, then hybridize and process the GeneChips. We will store the data on our firewall-protected server. We will analyze the controls and perform an initial, basic analysis of the results using GeneSpring, SAM or other appropriate software and present the user with a publication-quality figure. More intensive or customized data analysis is available for an additional charge.

Besides the complete Affymetrix system, the KUGR facility has a Nanospec spectrometer for quantifying RNA in small volumes, and an Agilent Bioanalyzer for rapidly analyzing the quality and quantity of RNA or DNA samples.

How much will an Affymetrix experiment cost?

For UNM users, Affymetrix GeneChips cost approximately $450 each. The additional reagents and kits required for the analysis cost $150 to $250 more per GeneChip, depending on the quality and quantity of RNA that is available. Additional or customized data analysis may cost $200 to $300 more. A typical experiment with 10 GeneChips (5 samples, in duplicate) will most likely cost $6,000 to $8,000.

Do I really need duplicates?

Yes! Microarray experiments generate large, complex data sets. Having replicates greatly enhances the quality of the data analysis. Duplicate samples are good. Triplicates are best. Replicates really are necessary to get good results that are meaningful and worth the cost.

How good is the Affymetrix data?

The newest Affymetrix GeneChips have probe sets for more than 50,000 genes. These are the fourth generation GeneChips and produce outstanding data with very low background. The correlations between independent replicates are routinely better than 0.97. In general, the Affymetrix system produces data that is very high quality and very reproducible, and the results are easily confirmed by Northern blot or real-time PCR. Affymetrix sells GeneChips for use with human, mouse, rat and many other types of samples. More information is available from the KUGR staff or at http://www.affymetrix.com

What controls are used in the analysis?

The quality of the starting RNA samples will be confirmed using the Agilent BioAnalyzer. We recommend adding internal "spiked" controls to the RNA samples before cDNA synthesis. The control RNAs come with the Affymetrix kit and provide a means of controlling for variations in cDNA synthesis or probe labeling. A biotinylated oligonucleotide is always included as a hybridization control. After hybridization, several quality control methods, including evaluation of scaling factors, comparison of 5’ to 3’ labeling ratios, the percentage of ‘Present’ calls and the expression of internal and ‘spiked’ controls, will be used to validate the quality of the microarray data. Any samples that fail to pass the stringent quality control measures will be identified and eliminated before the analysis.

How will the data be analyzed?

Initial data analysis and scaling will be peformed with the Affymetrix GCOS software. Advanced data analysis will be performed using GeneSpring software (Silicon Genetics), which the KUGR staff has used extensively. Advanced filtering and statistical tools will be used to identify genes that are reproducibly up- or down-regulated in response to the treatments.

What is the ‘day effect’?

There is a possibility that the complex process involved in generating microarray data will introduce some systematic variation (‘day effect’). We recommend that researchers perform completely independent replicate experiments. For example, prepare one complete set of controls, treated and untreated samples one day. Then prepare a completely independent and complete set of replicate samples on a different day. Do not, for example, prepare and run all the control samples together and then all the treated samples together, which amplifies the 'day effect'. Researchers are urged to consult with the KUGR staff before beginning the experiment to get tips on experimental design.

Who will help me analyze my microarray data?

The Technical Director of the KUGR Facility, Dr. Gavin Pickett (ggpickett@salud.unm.edu) is an expert at analyzing Affymetrix GeneChip data using GeneSpring, a high-end microarray data analysis software package from Silicon Genetics. The facility Director, Dr. Scott Ness, has analyzed results from hundreds of microarray experiments, has written review articles and book chapters about microarray data analysis, and is available for consultation. The KUGR staff is dedicated to helping users get the most out of their data and we have extensive experience analyzing Affymetrix data. We can also put researchers in contact with biostatisticians, computing and database experts and others who can help with more difficult problems.

How can I verify my microarray results?

Microarray results can be verified by several methods such as Northern blots or quantitative real-time PCR. Applied Biosystems sells "Assays on Demand" pre-made real-time PCR probe sets for most genes on the Affymetrix GeneChips. http://home.appliedbiosystems.com

What other services does the KUGR facility provide?

In addition to the Affymetrix service, we are currently amplifying and printing custom glass slide spotted arrays. We already have high quality custom microarrays for the analysis of yeast gene expression and we are developing assays for detection of alternative RNA splicing in human genes and high-throughput analysis of single nucleotide polymorphisms (SNPs) in human samples. Our instrumentation for high-throughput sample handling and preparation of custom spotted arrays includes a Bio-Tek Instruments Precision 2000 Pipeting station, three MJ Research PTC-225 DNA Engine Tetrad Thermal Cyclers set up for processing of up to 12 x 96 simultaneous PCR reactions, a Qiagen BioRobot 3000 for automated purification of up to 4 x 96 DNA samples every 2 hours and a BioRobotics MicroGrid II DNA spotter for automated preparation of up 120 custom glass slide microarrays per run. The facility has a collection of over 50,000 sequence-verified human cDNA and EST clones (Research Genetics), an Agilent BioAnalyzer 2100 for rapid analysis of small RNA and DNA samples, as well as –80°C freezers, UV crosslinkers, microwave oven, high capacity Speed Vac, centrifuges, electrophoresis and gel documentation equipment, etc, required for high throughput amplification, purification and analysis of PCR products. If you have a need for custom arrays or high-throughput, microplate fluidics, please contact us about pricing and availability information.

What information should I include in my grant application?

The facility Director, Dr. Scott Ness (ness@unm.edu), can provide letters of support and advice about how to describe the KUGR Facility and potential microarray experiments in grant applications. Dr. Ness has served on numerous NIH, ACS and DOD study sections and has reviewed many microarray-based grant applications. His own funded grants have microarray experiments in them. He can provide help with writing sections of your grant regarding microarrays and can point out potential pitfalls and things to avoid.

Things to avoid in grants containing microarray experiments:

The easiest way to criticize a microarray experiment is to describe it as a fishing expedition. Here are some things you should definitely avoid.

1) Do not propose to characterize genes that you have not yet identified. If you have no preliminary microarray data, you don't know what genes or how many genes you will find. However, they will likely number in the hundreds. Simply saying that you will pick some interesting genes to study is a quick way to get a bad score on your grant. If possible, your experiment should test a hypothesis. For example, you might make the hypothesis that certain genes (e.g. apoptosis genes) will get induced. Then you can propose to use microarrays to test that (and propose real-time PCR or Northern blots as back-up methods). That way you can test a hypothesis, propose expected outcomes and controls (e.g. genes that should go up and down), which is a much better way of doing a microarray experiment (or any other experiment). Simply going fishing for genes is a bad approach and always draws the ire of the review committee.

2) Simply saying that you will use some software program to analyze the data or group the genes into pathways is also going to get you in trouble. Microarray data can be extremely complex, and will require statistical methods for analysis. The pathway data that is known is woefully incomplete. Most genes are not in the pathways, anyway. You will need a well-planned approach for analyzing the data. You should have a way of telling whether the experiment worked or not (e.g. did the expected apoptosis genes get activated?).

3) The microarray experiment should not be merely a paragraph at the end of one of your aims. Whatever you do, absolutely do not add a microarray experiment to the end of a grant application as something that you "will also do". Microarray experiments are big, expensive and complicated and they can't be done as an afterthought. Many, many grants have a one-paragraph description of the microarray experiment that the researchers will also do. That is a lightning rod for criticism from the reviewers.

4) If you are looking for genes, you should be looking for them for a reason. Don't just propose to look for regulated genes without proposing to do something with them. Finding the genes that go up and down is not a significant enough goal. You need to be looking for genes with some purpose in mind (e.g. some hypothesis to be tested).