Jump to content

  • Log in with Facebook Log in with Twitter Log in with Windows Live Log In with Google      Sign In   
  • Create Account

Submit your paper to J Biol Methods today!
Photo
- - - - -

ELISA - calculate diluted concentration


  • Please log in to reply
6 replies to this topic

#1 icarus.rachel

icarus.rachel

    member

  • Active Members
  • Pip
  • 5 posts
0
Neutral

Posted 13 September 2009 - 07:39 AM

Hi, when the standard curve is already plotted, how to calculate the concentration from a 1:10 diluted samples? is it just multiplied by 10 as final concentration? but i cannot get the consistent result when i'm doing 1:2, 1:4, 1:8. Please suggest. Thank you.

#2 Gerard

Gerard

    Veteran

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 124 posts
0
Neutral

Posted 13 September 2009 - 09:23 AM

1:10 dilutions are just to be multiplied by 10.
Not lineair results in serial dilutions can have several causes.
First are you sure you used the correct program to plot the standard curve?
Ockham's razor
Pluralitas non est ponenda sine necessitate
-- "You must assume no plural without necessity".

#3 icarus.rachel

icarus.rachel

    member

  • Active Members
  • Pip
  • 5 posts
0
Neutral

Posted 13 September 2009 - 05:31 PM

i use the Ln(Absorbance) against Ln(conc) and generate a equation in excel. It looks good when compared with the standard.
So could you give some advice on calculate the concentration? instead of 1:10000, I make 1:10 by mistake.

Thank you very very much. :)

#4 Gerard

Gerard

    Veteran

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 124 posts
0
Neutral

Posted 13 September 2009 - 11:19 PM

If the absorbance of the sample is between the lowest and highest absorbance reading of the standard curve it is correct to multiply the result by 10 but 1 don't know if the elisa is valid for this difference in dilution of the sample.
As the standard curve concerns you know that in elisa the standard curve has a sigmoid shape that not kan be described as a liniair function
Ockham's razor
Pluralitas non est ponenda sine necessitate
-- "You must assume no plural without necessity".

#5 sgt4boston

sgt4boston

    Veteran

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 276 posts
1
Neutral

Posted 14 September 2009 - 09:43 AM

If plot is correct; %CV of standards OK...sounds like matrix effect to me.
When you run the next assay dilute both the sample (high) and the highest standard in parallel. The standard should dilute linearly. If this is so and the sample does not then it is a problem of the assay. Are you following the manufacturer's protocol for diluting high samples? Usually they have you dilute in buffer or using the "0" standard. The standards should be the same matrix as the sample or as close as possible.
Another option is some type of heterophile ab in your samples which give you lower results for your sample and higher results for the dilutions.

#6 aliu

aliu

    member

  • Members
  • Pip
  • 4 posts
0
Neutral

Posted 19 November 2009 - 07:40 PM

For elisa analysis, I would recommend using the 4-PL or 5-PL non-linear regression models as that would fit your data quite nicely.

To account for your diluted samples, you can simply multiply the 1X concentration by 10. You say you have inconsistent results though. Do you have replicate groups for both your standards and unknowns? If not, I would definitely recommend redoing the experiment with replicates as that will tighten up your standard curve. Also, do you use weighting in your concentration calculation to offset heteroscedasticity. Here are some useful tips for elisa analysis that includes a discussion on that.

MasterPlex ReaderFit is piece of software that is dedicated to Elisa analysis and takes into account dilution factors, heteroscedasticity, and also provides the 4-PL and 5-PL model equations as well as some others. You can try it out for free to see if it is the right solution for your needs.

As you probably have guessed by now, I am a representative of the company that designed this software. In fact, I am the tech support supervisor for the software. If you would like, we can even take this offline and I will go over the live analysis of your data with the software (with no strings attached of course as I am not a sales person) :D I am only looking to solve problems with our solutions.

Anyway, I hope this information helps.

#7 WolfeMD

WolfeMD

    member

  • Active Members
  • Pip
  • 18 posts
0
Neutral

Posted 13 December 2009 - 03:20 PM

Whatever program you use the key think you will want to look for is the R-square of your regression. That is, you are making a scatter plot of OD vs. Concentration, hopefully with at least two reps of each dilution per plate. You regress a line either linearly or in some of form through those points. The r-square tells you how much of the variation between dilutions is explainable by the regression line. I generally consider r-square of 0.95 (95% of variation explained) and above a reliable standard curve.




Home - About - Terms of Service - Privacy - Contact Us

©1999-2013 Protocol Online, All rights reserved.