Looking for experimental planning information? Browse the topics listed.

For additional information, browse the answers to frequently asked questions (FAQs) by miRNA researchers. Need assistance performing miRNA profiling experiments? Try using Invitrogen's miRNA Profiling Service.

Sample Size

  • We recommend running each sample at least twice, once with each dye. This helps reduce the potential of any dye effects with your sample for individual markers, as well as provide replicates for analysis.
  • We recommend starting with250 ng-1 µg of total RNA . This would be  500 ng - 2 µg of total RNA for each sample to run one replicate of the sample in both colors. This If you are using the NCode™ Amplification Kit, you can use as little as 50 ng of total RNA for one color.

Special Equipment

Data Analysis Methods

Three methods are typically used for normalization of 2-dye expression profiling microarray experiments:

  • Latin Squares/Loop Design/Dye Swap (used in Invitrogen’s NCode™ profiling services – see recommendations below)
  • M vs. A, Lowess Normalization
  • Quantile Normalization

Refer to the Epigenetics Learning Center for a detailed description of the three methods.

Use of the Latin Squares/Loop Design/Dye Swap Methodology with NCode™ Products
We recommend using the Latin Squares/Loop Design/Dye Swap Model described by Kerr, et al because this method provides the reliable statistical data with the smallest number of microarray chips.

Considerations:

  • Combining data from multiple miRNA microarrays requires scaling and/or multi-array normalization methods.
  • As a general rule, we recommend that with k tissues, to use a k-multiple of arrays so that a complete loop design with replicates can be conducted. With this design, i.e. k tissue, using say n*k arrays where (n can be 1 or 2 or 3 or...) will then have 2*n replicates of each tissue in the experiment.
  • The loop design format makes the modeling relatively straight forward. For example, with 3 samples the experimental design would be as follows:
    • the first array sample A would be labeled with Alexa Fluor® 3 and sample B with Alexa Fluor® 5
    • the second array sample B would be labeled with Alexa Fluor® 3 and sample C with Alexa Fluor® 5
    • the third array sample C would be labeled with Alexa Fluor® 3, sample A with Alexa Fluor® 5

This experimental design make a loop design without a chip replicates, though each tissue will be replicated twice, once with each Alexa Fluor® dye.

  • Using this array layout enables comparison of each test sample with the reference sample AND comparison of each test sample with any other test sample.
  • Using a reference design, where every sample is compared to reference sample, would essentially be a Dye Swap model (a loop design with 2 samples) for each sample. This method requires 2*k chips for k tissues (not including the reference sample). This design requires twice the number of chips of a loop design.

Figure 1 Example three and ten array experiments.
T refers to tissue sample, and each array is run with Alexa Fluor® 3 on the left and Alexa Fluor® 5 on the right.

Three Array Experiment
with 3 tissue samples , no replicates

Array 1: T1 - T2
Array 2: T2 - T3
Array 3: T3 - T1
Ten Array Experiment
with 5 tissue samples with 2 replicates

Array 1: T1 – T2
Array 2: T2 - T3
Array 3: T3 - T4
Array 4: T4 - T5
Array 5: T5 - T1
Array 6: T1 - T5
Array 7: T5 - T4
Array 8: T4 - T3
Array 9: T3 - T2
Array 10: T2 - T1

Statistical consulting for experimental design and analysis is available for a fee. Contact Invitrogen Custom Services for more information.

Data Analysis Tools

There are several open source, freeware, and commercial data analysis software available. Four leading options are described:

NCode Profiler™ (Invitrogen)
Free experimental design and analysis software for two-dye expression profiling miRNA microarray experiments.

R (open source)
Free, open source software for statistical computing and graphics based on the statistical programming language S hosted by the R Project for Statistical Computing.

Bioconductor (open source)
Free, open source software for the analysis and comprehension of genomic data hosted by the Computational Biology Group in the division of Public Health Sciences at the Fred Hutchinson Cancer Research Center. It includes a collection of scripts and functions for R developed by the microarray user community (see above).

GeneSpring™ (Agilent Technologies)
A commercial visualization and analysis software developed specifically for analyzing of gene expression microarray data. A two-week trial version is available for download online.