Hello!
I've been tasked with comparing the DNA quantification abilities of the Qubit Fluorometer and our lab's Filtermax F5 Microplate Reader. We've been using the fluorometer so far but with many samples coming in we're looking to use the microplate reader to speed things up.
I prepared a series of standards using the 10.0 ng/uL standard provided in the Qubit 1X dsDNA HS Assay Kit to a final volume of 210 uL each. The reason for making my own series of standards was to achieve a more accurate calibration curve for samples with lower concentrations. I also prepared three sample tubes with 2 uL of sample and 198 uL of Qubit working solution for a final volume of 200 uL each.
I prepared the samples in microtubes that fit into the fluorometer, read the samples in the fluorometer and then transferred the solution from each tube to a black 96-well plate to be read in the microplate reader. I set the excitation to 485 nm and the emission to 535 nm.
I'll attach a photo of my results. The fluorometer provides its own concentration estimation directly when samples are read. For the microplate reader, I constructed a calibration curve by plotting the fluorescence against the concentration of the standards and then used the equation of the line to estimate the concentrations of the samples. I don't know why the microplate reader is estimating a concentration that is nearly double what the fluorometer estimated?
I don't have an easy answer, but I can offer a few relevant observations from my experience.
One is that fluorimetry is fraught with artifacts and confounders. Fluorescence yield is affected by common buffer species and refraction by the cuvette. It's more sensitive than absorbance but noisier and less linear. That's why most people estimate DNA concentration using absorbance (OD260), which is less sensitive but more robust.
Second is that I've seen amazingly different DNA concentration numbers between multiple devices. I tend to trust simple instruments. My sense is measuring OD on a 1mL sample is more reliable than measuring OD in 10uL cuvette.
A related third observation is that scientists usually pick a single assay platform and stick to it. There is always some systematic error. But when you calibrate your assay (sample prep, instrument, cuvette) with a curve, you can count on it for that assay. If you change platforms you essentially change assays, and you need to recalibrate. That's one reason results are sometimes difficult to repeat between different labs.
Finally, does a factor of 2 on a single measurement really matter for you? For a good curve you should measure at least triplicate samples comprising independently prepared dilutions. Expect variance and do the statistics. Use proper significant figures (probably two) and you may find a factor of two is acceptable.
- Badges
- Science topic
- Similar topics
- Instrumentation Engineering
- Instruments