Performance and application data for Qubit Fluorometric Quantification

Sample data on DNA quantification, RNA quantification, RNA integrity and quality (IQ), protein quantification, and endotoxin detection demonstrate that data generated with Qubit Assays on Qubit Fluorometers is accurate, precise, sensitive, and selective.


DNA quantification data

Accuracy and precision

Accuracy and precision of Qubit Fluorometers. Accuracy was assessed as average deviation from true concentration using the Qubit dsDNA HS Assay, while precision was evaluated as coefficient of variation (CV) across replicates of various concentrations using the Qubit dsDNA BR Assay. The low deviation percentages demonstrate the accuracy of the Qubit Flex and Qubit 4 Fluorometers, while the low CV percentages demonstrate their precision, especially the Qubit Flex model. Of the three instruments tested, the competitor’s instrument had the lowest accuracy and precision.

Accuracy and sensitivity

Qubit dsDNA accuracy and sensitivity vs UV absorbance. Ten replicates of lambda DNA at known concentrations from 0.01 to 10 ng/μL were assayed using the Qubit dsDNA HS Assay on the Qubit Fluorometer (blue bars) according to the standard kit protocol. The same concentrations of DNA were measured via UV absorbance in 10 replicates using a microvolume spectrophotometer (blue bars), and results were compared for both accuracy and precision. Each bar represents the average of 10 replicates, while error bars (almost indiscernible for Qubit measurements) represent their standard deviations. The x-axis scale represents the known concentrations of DNA in the starting samples, before dilution in the Qubit assay tubes, while the y-axis scale shows the measured concentrations. The Qubit Fluorometer measurements were more accurate (y » x), sensitive (low concentrations, inset), and precise (much smaller error bars) than the UV absorbance measurements.

Sensitivity and selectivity

Qubit dsDNA HS Assay sensitivity and selectivity

The Qubit dsDNA HS Assay reported accurate, linear results for known, ascending quantities of dsDNA (green circles), even in low amounts (inset, enlarging the very low end of the scale). These DNA-specific measurements were only slightly affected by RNA added alone (red triangles) or in combination with DNA (blue squares).

A) Qubit 1X dsDNA HS Assay

B) Qubit 1X dsDNA BR Assay

Selectivity of Qubit 1X (A) dsDNA HS and (B) BR assays

Plots show the quantification results for samples seeded with known quantities of DNA and RNA vs expected values. Circles represent samples with 10 μL of DNA plus 190 μL of working solution, at varying concentrations. Squares represent samples with 10 μL of RNA and 10 μL of DNA plus 180 μL of working solution, at varying concentrations. The closeness of the circles and squares—in some instances almost indistinguishable—demonstrate that these two dsDNA assays are minimally affected by the presence of RNA.


RNA quantification data

Accuracy and selectivity

Qubit RNA and microRNA assays

Qubit RNA and microRNA assay accuracy and selectivity. Ribosomal RNA (rRNA) at the concentrations listed on the x-axis was added to samples containing 2 μg/mL siRNA. The mixtures were then assayed using the Qubit microRNA assay, the Qubit RNA assay, and the NanoDrop A260 assay, which quantifies using UV absorbance. Results from eight replicates were averaged, with standard deviations shown. The NanoDrop instrument’s (purple bars) detection limit for total RNA concentration is 1.5 μg/mL, affecting its accuracy and precision at low concentration levels. The Qubit RNA assay (red bars) accurately quantified the rRNA concentration over a broad scale. The Qubit microRNA assay (blue bars) accurately quantified the 2 μg/mL siRNA concentration, while accuracy was mildly affected as rRNA increased from 2 to 5 times that amount. The blue and red trendlines indicate the actual concentrations of siRNA and rRNA in the samples, respectively.

Accuracy, precision, and sensitivity

Qubit microRNA assay accuracy, precision, and sensitivity. Six experiments were run to test the accuracy and precision of the Qubit microRNA assay with pure siRNA; results of a typical experiment are shown here. In this experiment, eight replicates of GAPDH siRNA in known concentrations from 0.1 μg/mL to 12 μg/mL were measured using the Qubit microRNA Assay Kit with the Qubit Fluorometer and using A260 measurements on the NanoDrop ND-1000 Spectrophotometer. The detection limit for the NanoDrop instrument is reported to be 1.5 μg/mL; lower siRNA concentration results are shown for comparison. Accuracy for the Qubit microRNA Assay was within 15% of expected. CVs (1 standard deviation/average of 8 data points) were 0.63% to 2.9%. For all six experiments, the Qubit microRNA assay accuracy was within 15% of expected, and CVs were <5%.

Accuracy and breadth

Qubit microRNA assay accuracy and breadth. Concentrations of pure siRNA and miRNA samples were determined by optical density on a PerkinElmer® spectrophotometer at concentrations yielding an A260 of 0.3–0.6. Samples were then diluted and tested with the Qubit microRNA Assay at four different known concentrations. Results show accurate quantification of (A) four different single-stranded siRNA molecules and (B) two double-stranded siRNAs and two double-stranded miRNAs.


RNA integrity and quality (IQ) data

Selectivity and validity

Selectivity of the RNA IQ assay reagents for large and small RNA. Triplicate samples containing 100 ng/µL rRNA (E. coli) and varying amounts of siRNA (0–50 ng/µL) were assayed with the Qubit RNA IQ assay on the Qubit 4 Fluorometer. Graphs show (A) relative fluorescence units (RFUs) and (B) IQ scores for these samples. The data shows that the dyes are specific to large (like rRNA) and small (like siRNA) RNA respectively, and that IQ scores are strongly correlated with the percentage of large RNA in the sample.

rRNA degradation by RNase A over time, measured by the Qubit RNA IQ Assay. Triplicate samples of 100 ng/µL rRNA solutions were incubated with different doses of RNase A in the final assay solution containing multiplexed dyes and assay buffer. (A) Higher doses of RNase A caused faster degradation of RNA over 60 minutes as reflected by its RNA IQ score. (B) Exposure to 10 fM RNase A caused a gradual decline in large RNA fluorescence over the hour and a corresponding increase in small RNA fluorescence. (C) With exposure to 100 fM RNase A, the changes occurred more quickly.

Validation against RT-qPCR

RNA IQ is consistent with RT-qPCR results while Bioanalyzer™ RNA integrity number (RIN) is not. Total RNA (isolated from human liver) was heat-treated at 75°C for varying durations and analyzed using the Agilent™ Bioanalyzer™ system and Qubit RNA IQ Assay. RT-qPCR analysis was also performed using RETROScript reverse transcriptase and TaqMan hHIF1α and hGAPDH assays. (A) Data from the Bioanalyzer system show rapidly decreasing rRNA peaks over time, suggesting deteriorating RNA integrity. (B) Comparison of RIN (black dots) and RNA IQ (gray triangles), including more detailed results at the 40 min time point, shows disagreement, with RIN decreasing rapidly but RNA IQ largely stable over time. (C) RT-qPCR results were also stable over time, in agreement with RNA IQ but not RIN.


Protein quantification data

Accuracy and precision

Accurate determination of protein load from complex protein mixtures. The Qubit Protein BR Assay and a standard Bradford assay were used to determine the protein concentration of lysates from several mammalian cell types: 293T, A549, HepG2, HeLa, and iPSCs. Lysates were separated on an Invitrogen NuPAGE 4–12% Bis-Tris Mini Protein Gel and labeled with No-Stain Protein Labeling Reagent. (A) Gel image was acquired on the iBright FL1500 Imaging System and (B) normalization factors were determined using Invitrogen iBright Analysis Software. The coefficient of variation (CV) of the Qubit Protein BR Assay was appreciably lower than that of the Bradford assay.


Endotoxin detection data

High sensitivity and broad range


The Qubit Endotoxin Detection Assay utilizes a streamlined single-incubation step workflow and offers a broad detection range of 0.01–1.0 EU/mL when using 50 µL of sample. The assay can accurately detect up to 10.0 EU/mL using variable sample inputs.  Simply select the Endotoxin icon from the home page of the Qubit Flex Fluorometer. Use the provided Endotoxin standard to generate a 4-point calibration curve and measure up to 8 samples at a given time using the Qubit Flex Pyrogen Free Assay Tube Strips (Cat. No. Q32893).

Automatic calculations

Expected concentration (EU/mL) 0.010 0.050 0.100 1.00

1.00

5.00

10.00

Average measured concentration (EU/mL)

0.0098

0.045

0.099

0.98

0.97

4.96

10.0

CV

5%

2%

8%

5%

4%

6%

10%

Relative error

2%

11%

1%

2%

3%

1%

0%


When paired with the Qubit Flex Fluorometer (Cat. No. Q33327), calculations are performed automatically reducing the potential for error. The Qubit Flex Fluorometer automatically calculates the correlation coefficient using log-transformed linear regression as described by the U.S. Pharmacopeia. Then the Qubit Flex data is analyzed using log transformed data and a background corrected quadratic fit. Using the Qubit Flex, 5 µL and 50 µL samples generated accurate and reproducible results. Five microliter samples (gray) had an average CV <7% and average relative error <5%.  Fifty microliter samples had an average CV of 5% and average relative error <5%.