Standard curve standardization - variation between runs (Apr/24/2009 )

Hi folks,

I need to run a standard curve to calculate the viral load of the virus I work on, and I'm becoming a little confused as how to standardise between runs.

We have a Roche LC480. This means I only need to run the standard curve only once, save it to the machine, and then on subsequent runs I include a cDNA sample from the dilution series to recalibrate the curve. Obviously when I run out of samples from my dilution series I need to remake a dilution series and rerun the curve.

I've noticed that on subsequent runs, my frozen standards do not always match the dilution they are supposed to, so the the curve is slightly off. Does anyone know how to correct for this and standardise between runs? When making my dilution series I aliquot the samples to minimize free/thawing. Should I include another sample and always run this on subsequent plates to observe any variation? If so, can I use the same primers as with the curve or should I amplify another gene?

Any suggestions, comments gratefully received

Thanks,
Bunsen

I an not familiar with your methodology but can suggest the following for standardization between runs/days/curves etc.

Use the midpoint of your reference curve and normalize the results from your other runs based upon that number. That is there will be a 'factor' which you multiply all your subsequent run values by to normalize them to the one point. That point will be constant across all runs. The curves if similiar in shape will then align on top of one another.

A second question is just how far 'off' are you? can the variation just be due to instrument variability and not the samples? Do a precsion test.

I am still learning about this. But as far as I understand, I think you will need to use your standards every time. it might be difficult to compare between different runs.

Bunsen Honeydew on Apr 24 2009, 04:07 PM said:

Hi folks,

I need to run a standard curve to calculate the viral load of the virus I work on, and I'm becoming a little confused as how to standardise between runs.

We have a Roche LC480. This means I only need to run the standard curve only once, save it to the machine, and then on subsequent runs I include a cDNA sample from the dilution series to recalibrate the curve. Obviously when I run out of samples from my dilution series I need to remake a dilution series and rerun the curve.

I've noticed that on subsequent runs, my frozen standards do not always match the dilution they are supposed to, so the the curve is slightly off. Does anyone know how to correct for this and standardise between runs? When making my dilution series I aliquot the samples to minimize free/thawing. Should I include another sample and always run this on subsequent plates to observe any variation? If so, can I use the same primers as with the curve or should I amplify another gene?

Any suggestions, comments gratefully received

Thanks,
Bunsen

Just a thought, but does this machine automatically set the threshold? If it's setting the threshold to a different level for each run, that would affect your Ct values.