Microarray

At the discretion of our clients, the microarray team can be involved in every step of a microarray project, and is equipped to run all major array formats including Illumina, Affymetrix, and Agilent.  We help with study design, preparing samples, we run the hybridizations, and can analyze the results.

GTAC Microarray FAQ's

Click on Any Question to see the answer

I wish to ship my samples to GTAC. What is your shipping address?


Please ship samples on dry-ice to:
Genome Technology Access Center/ Cortex Building
Department of Genetics
Washington University School of Medicine
4444 Forest Park Parkway
St. Louis, MO 63108


What microarray platform should I use for my experiment?


We offer all major brands of microarrays from Affymetrix, Illumina, and Agilent, and we can help you to choose a suitable platform for your experiment, taking into account the following considerations:

  • your study goal
  • sample size
  • sample quality
  • available funds
So give us a call or email, and we can discuss it with you!


What are the differences between 3’microarrays and whole transcript microarrays?


3’ microarrays have probes predominantly targeting the 3’end of transcripts.  While these arrays usually require high-integrity RNA samples, GTAC does provide some specialized protocols in which Agilent and Illumina arrays can be used with degraded RNAs.

Whole-Transcript arrays have probe sets which hybridize to loci throughout the entire transcript, and these arrays would work on both high quality and partially degraded RNAs.

Examples of 3' microarrays are:

  • Illumina Human_HT12
  • Mouse_6 and Mouse_8
  • Agilent 8X60K, 4X44K, 8X15K
  • Affymetrix Human U133plus2 and Mouse_430_2
Examples of whole-transcript include:
  • Affymetrix Gene ST arrays (1.0 and 2.0)
  • Affymetrix Exon 1.0 ST arrays.


Can I prepare my own amplified, labeled and fragmented samples for hybridization?


Yes, this has been done before with good sucess.  We recommend following all protocol guidelines.
However, GTAC cannot guarantee microarray results for samples processed outside of GTAC.


Are there a minimal number of samples I should include in my experiment in order to obtain statistically significant data?


There is no simple answer to this question.  Detecting a significant change in a gene’s expression depends on three main factors: the actual magnitude of the differential expression, the amount of background noise, and the number of replications. Biological replicates increase the statistical power of the analysis, allow for resolution of any outliers, and lend more confidence to interpretations of transcriptional change.  While minimum of three (3) biological replicates is recommended, there may be conditions in which biological variation may be so extensive that more replicates should be considered.   For example, tumor tissue from different individuals might vary significantly in the environment in which it formed and its cellular characteristics.   That said, we recognize controlling costs is an essential element for investigators. However, weighing the usefulness of the data vs. the cost of results should be a prominent factor in deciding on the design of a project. Perhaps fewer conditions with more replicates may be an acceptable strategy to get the most out of a project.


Is it advisable to combine data from experiments done at different time or using different platforms?


Combining microarray data from the same platform is easy to do, but does introduce batch effects (biological, lot, reagents, etc), reducing sensitivity BUT increasing specificity (more false negatives but less false positives).  
 
Cross-platform comparisons are much more difficult.  They require mapping files which are usually created from common annotation such as gene symbol.  Many probes simply do not map to the other format and the ones that do may vary in length, specificity and location in the transcript, all of which may reduce the meaningfulness of the data.  It can be done, but it is difficult and limits the amount of information that can be derived from the project.


Will GTAC help me analyze my data?


Yes, the GTAC provides analytical support, such as advanced statistics and pathway analysis,  to its customers on a fee-for-service basis.


How do you deal with batch effect from individual chips or from individual hybridization done at different times?


Batch effects can be mitigated using statistical analysis, such as ANOVA, as long as they do not coincide with the experimental conditions.  For multiplex chips (multiple arrays per chip) such as Agilent and Illumina, there is usually a small amount of chip to chip variation.  If samples from one condition are confined to one multiplex chip, then it is impossible to tell if the apparent variation is from biology or batch.  We address this by careful distribution all conditions across all the chips used. For single cartridge arrays (i.e. Affymetrix), the vendor has implemented an algorithm called RMA (Robust Multiple array Average) to normalize the chip-to-chip variation.


How do I retrieve my microarray data?


Affymetrix data is uploaded and archived to the Bioinformatics server, an email is automatically sent to notify the Investigator, and data is available to the Investigator via download from the Bioinformatics website. If you have questions or problems with downloading, contact Bioinformatics Help at help@bmi.wustl.edu.

Illumina and Agilent data is stored locally and an email is sent with FTP download instructions.

GTAC requires that you download and store your data within 10 days of the initial export.  It is a good idea to store your data in a backed-up, secure location.  After 10 days and for up to 6 months, GTAC can retrieve and re-export data; however, additional storage/retrieval fees may apply.


What files will I receive upon download of my data?


For Affymetrix arrays, you will receive:

  • RMA-processed and quantile normalized .txt data in spread sheet format
  • raw chip data files (.CEL) that are generated by the scanner.
  • n.b. For Affymetrix Gene ST and Exon ST arrays, there is no detection p-value for each gene when data are summarized at gene-level, but a detection p-value IS calculated for each exon probe set when data are summarized at exon-level.
For Agilent:
  • feature (probe) extracted .txt data for each sample.  These data are quantile normalized without detection p-value.

For Illumina:
  • normalized data
  • non-normalized data
  • Data will be in both spread sheet and Partek-ready formats. Illumina data have detection p-value attached for each data point at both probe- and gene-level.


I'd like to acknowledgement GTAC for its great work! What's the best way to do that?


If data generated by GTAC is used in an abstract, publication or other formal communication, please use the following acknowledgement:
 
We thank the Genome Technology Access Center in the Department of Genetics at Washington University School of Medicine for help with genomic analysis. The Center is partially supported by NCI Cancer Center Support Grant #P30 CA91842 to the Siteman Cancer Center and by ICTS/CTSA Grant# UL1 TR000448 from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH), and NIH Roadmap for Medical Research. This publication is solely the responsibility of the authors and does not necessarily represent the official view of NCRR or NIH.
 
Further important considerations:

  • NIH Public Access Policy: Publications supported in part by NIH funding must adhere to the Public Access policy.  Before signing copyright agreement for any publication, be sure the journal supports this policy.  Please see the WU Becker Library website for additional information https://becker.wustl.edu/services/scholarly/nihpolicy.html) and the process for compliance with this policy, including demonstration of compliance.  For assistance, please contact Cathy Sarli (sarlic@wustl.edu).
  • Please cite the appropriate CTSA grant number: NIH requires acknowledgement in all publications and projects which resulted from research benefitting from an ICTS core or service:  “This publication was made possible by the NIH CTSA Grant # UL1 TR000448 from the NIH National Center for Research Resources."
  • ICTS CRTC Scholars/Trainees should reference the applicable NIH linked-award number noted below:
  1. KL2 TR000450 - ICTS Multidisciplinary Clinical Research Career Development Program
  2. TL1 TR000449 - ICTS Clinical Research Predoctoral Training Program


What is the sample to data turn-aorund time?


For <48 samples our average turnaround time is 3 weeks.  For larger projects we will provide an estimate on a case-by-case basis.