As you all know, the concentration of various biological substances are being quantified through calorimetric assays. Most of the protocol, the given formula to calculate the concentration of unknown substance is = Test OD/Std OD * Std Concentration. In my case, i am running standard at 4 or 5 different concentration and plotting standard curve for each assay. While calculating concentration of unknown, i have to take OD of which standard concentration as there is 4 or 5 standard OD. It is noteworthy that, i get different results for unknown concentration while taking different Standard OD.
I need to be clarify two things.
First, is there any cut-off value of R2 pertaining to standard curve, i.e. upto certain level, standard curve can be accepted or must not be considered.
Secondly, is there any ideal formula to calculate the unknown concentration. Because i observed that, selection of standard OD have impact on unknown concentration. Currently, i used to select the standard OD only by arbitrary.
Note: Please find the attached excel file, in which i have given the real data of one calorimetric assay. In this assay, i run standard (malondialdehyde) at 4 various concentration, i.e. 2 nanomolar, 4 nanomolar, 6 nanomolar, and 8 nanomolar. The R2 value of standard curve is 0.9993. I have got different results for unknown concentration while taking different standard OD.
Your R2 is a statistical measure of how close the data are to the fitted regression line given as a % of the response variable variation that is explained by a linear model. In your case your Std Curve is linear with an R2 of > 0.999 = > 99.9%! Thus I don't think you have an issue with your R2.
As for your quantitation, why don;t you use your trendline for quantitation instead of the single Std point? Your Std curve is linear with an excellent R2; thus I would use the trendline y= mx + b solved for your concentration rather than a one point quantitation:
x = (y-b)/m
with x as the unknown concentrations, m = 0.0038, y= measured OD, b = 0.045 (as given in your excel file).
This should be applicable as long as your unknown concentrations fall into your Std curve concentrations to assure linearity.
I would always prefer to quantify using the trendline. If you use one one Std you end up extrapolating from a single point without a second reference point, which then could give you the differences in concentration when using the different Std concentrations. Using the trendline you are covering a range of concentrations and the slope as well as your b.
Good luck!
The reason of the difference of your result comes from the fact that the response factor (OD divided by conc.) for each of your standard solution gives different value. The formula you have applied is based on the assumption that your calibration curve crosses zero point. But very often this is not a case. Thus you have to apply full standard curve equation because this covers all the bias related to the method.
Look into attached corrected Excel sheet (sheet2). Good luck
I agree with Daniela and Malgorzata, but in addition R2 cannot be used as a sole parameter to express the linearity. Please read the attach publication p 2233 (appendix A), additional parameter(s) was (were) needed.
