miRLink™ v11.0 Array Performance Specifications
Performance specifications of the miRLink Array Service reflect the robustness of
the CodeLink® platform and the proficiency of Asuragen's scientists.
|
Performance Specifications
|
|
Metric
|
Asuragen's method
|
|
Recommended Input
|
500 ng total RNA (no fractionation required); 1 ug for cell lines
|
|
LOD1
|
< 5 amol
|
|
Dynamic Range1
|
> 2.5 logs
|
|
Reproducibility (R2)2
|
0.99 (median CV < 3%)
|
|
Dose-Response Linearity (R2)1
|
0.99
|
|
Relative Accuracy1,3
|
1.12 ± 0.18
|
|
Specificity4
|
> 1 nt with < 10% cross-hybridization signal from highly-related, similar sequences
|
|
Sample QC
|
Incoming sample QC by Spectrophotometer, and TaqMan qRT-PCR.
If, for any reason, a qualified sample does not pass our process control specifications,
we will re-run the sample at no additional charge.
|
Footnotes:
1... Limit of Detection (LOD, Dynamic Range, Dose-Response Linearity,
and Relative Accuracy are all based on latin-square spike-in study in a complex
background of placenta total RNA processed according to the miRLink Array Service
protocols.
2... Reproducibility reflects linear R2 values between full-process
technical replicates.
3... Relative accuracy describes the degree to which the system responds
to linear changes in the concentration of spike-in controls. The dose response slope
measures the accuracy of reported fold-change relative to the known fold-change
of a serial dilution series of miRNA spikes. A slope of 1.0 indicates perfect relative
accuracy; slopes of approximately 0.8 are typical of microarray-based platforms
as they are known to compress data.
4... Specificity of miRLink arrays is based on performance assessment
on let-7 family members.
Linear Dynamic Range
Two process replicates are demonstrated here, illustrating the range of spikes
used in the Latin-square design (5 amol to 1.6 fmol). Note that the range defined
by the spikes brackets the biologically relevant range of the placenta total RNA.
*Calculations are for 500 ng of input mass with assumption of 5pg RNA/cell.
Outstanding correlation with qRT-PCR and other commercially available platforms,
even from fixed (FFPE) tissues
Eight FFPE blocks corresponding to four samples of two different tissue types (A
and B) from eight different patients were purchased from Proteogenex Inc (Culver
City, CA). All paraffin blocks were less than two years old. Total RNA was extracted
from each of these blocks by Asuragen’s FFPE RNA isolation services team. 500 ng
of total RNA isolated from each FFPE sample was analyzed with the Asuragen miRLink
Service; 200 ng of total RNA was analyzed using the Agilent V2 miRNA Array and 100
ng of total RNA was analyzed with the Asuragen DiscovArray Service.
A series of qRT-PCR reactions were performed using TaqMan microRNA assays#
specific for 6 different miRNAs following manufacturer’s instructions and using
an input of 15ng total RNA per reaction. Real-time reactions were run on the 7900HT
fast realtime PCR system# and initial analysis of data was done using
the SDS 2.3 software included.
The following figures demonstrate the remarkable correlation between the qRT-PCR
data and the data obtained from different array-based platforms with high statistical
significance (p-values as shown). These p-values demonstrate the overall reproducibility
of the new miRLink process.
In all figures, the array data shown are VSN normalized values and the TaqMan data
are ΔCTs.
Let-7 Family Specificity Study
Specificity (cross hybridization of probes) is an important metric when considering
microarray platforms. A microarray with adequate specificity should be able to distinguish
between highly correlated family members with over 90% sequence similarity. To test
the specificity of Asuragen’s custom miRLink v11.0 array, we examined the ability
of the array to distinguish several distinct miRNAs from the let-7 family. Each
of the eight members of the let-7 family was spiked individually (1 fmol) into 50ng
of yeast t-RNA and labeled and hybridized onto the microarrays. As a comparison
similar data were included for the Agilent miRNA Array platform as reported by Agilent
(Wang et al., RNA, 2007).
The figure shows probe–target sequence specificity of the human let-7 family. The
total signals reported by all the probes were normalized to the perfect match probe–target
hybrid for each microarray. The numbers represent percent of signal detected for
each spike as normalized to the perfect match probe-target pair. The colors represent
number of mismatches, as shown in the color key below the figure (green – perfect
match, yellow – one mismatch and so on).
Interested in Asuragen's other miRNA services? [Click here]
#...(Applied Biosystems, Foster City, CA, USA)