Hi all
This must have happened to others before. For example an IFN stimulated gene like MX1 goes from undetectable (>40 cycles) but then after IFN alpha stimulation the Ct is now 28. Both have similar reference gene expression (eg. HPRT) so no issues with RNA/cDNA it's just this cell type doesn't have much MX1 until after stimulation.
Now obviously that is a huge increase in expression but how can you show the data without a data point for your unstimulated? Changing cDNA amount, primers, etc does not affect signal. I would not think you can just give a Ct of 40 to your unstimulated to calculate ddCts so is there a way to still show the data? Maybe a way to show dCts only?
Thanks so much
It isn't, technically, a huge increase in expression, because 'increase' sort of implies there was expression previously, which there wasn't.
You literally have "no expression >> some expression", so fold changes are not only inappropriate, they're meaningless. And this is fine.
A Ct of 28 is still reasonably low, incidentally, so you could simply state the expression goes from 'undetectable' to 'low, but detectable', maybe in combination with some estimates of transcript number per ng of RNA (via standard curve).
If you're keen to plot this on a chart of some kind, you could always take multiple time points: expression will presumably go from 'zero' to some maximal value over a given time, so you could simply isolate RNA at two or more time points after IFN stimulation, then plot those. Perhaps call your maximum expression "1" and express everything else as fractions of that (your unstimulated will therefore be "0").
Remember: unless you're quantifying absolute transcript numbers, the Y-axis of a graph showing qPCR data is essentially arbitrary. All that matters is the relationship between the datapoints: the scale could be anything you like.
It is going to be hard to calculate fold change when your non-treated samples are undetectable( >40 cycles). You could plot dCts or even Cts.
It isn't, technically, a huge increase in expression, because 'increase' sort of implies there was expression previously, which there wasn't.
You literally have "no expression >> some expression", so fold changes are not only inappropriate, they're meaningless. And this is fine.
A Ct of 28 is still reasonably low, incidentally, so you could simply state the expression goes from 'undetectable' to 'low, but detectable', maybe in combination with some estimates of transcript number per ng of RNA (via standard curve).
If you're keen to plot this on a chart of some kind, you could always take multiple time points: expression will presumably go from 'zero' to some maximal value over a given time, so you could simply isolate RNA at two or more time points after IFN stimulation, then plot those. Perhaps call your maximum expression "1" and express everything else as fractions of that (your unstimulated will therefore be "0").
Remember: unless you're quantifying absolute transcript numbers, the Y-axis of a graph showing qPCR data is essentially arbitrary. All that matters is the relationship between the datapoints: the scale could be anything you like.
John nailed it.
To add: You may show and analyze the proportion of PCRs with detectable (and specific) amplification. You can use a binomial test to get the significance of your data (detected: yes/no) under the hypothesis that the proportion of "yes" is equal in both groups.
- Badges
- Science method