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
- - - - -

PCR analysis method- delta or delta delta?


  • Please log in to reply
4 replies to this topic

#1 Upa

Upa

    member

  • Active Members
  • Pip
  • 5 posts
0
Neutral

Posted 10 September 2010 - 08:19 PM

Hello,
I have been working on a project related to platelets RNA. So, I try to compare amounts of RNA of specific genes in platelets ( diseased vs normal, non-diseased patients). I extracted RNA from platelets using RNeasy and QiaShredder column. Then I measured the RNA concentration, and the samples were stored at -70C. Then, I performed qRT PCR with the RNA. I use ABiosystems SYBRGREEN, I calculated RNA total ng amount in each well ( so it is around 20 ng in each well) and ran PCR. I have lets say 2 samples, diseased and normal. Each sample is normalized (theoretically ) to a houskeeping gene, like actin b or GAPDH. So, I ran a sample against particular primers of the genes of interest in separate wells. Then I calculated average Ct for each primer (from diseased and normal sample separately) and normalized each primer's average Ct to the internal houskeeping gene. I used delta Ct method.I got huge stdev values... Should I use delta delta Ct method? ( I know it is used when treated vs non-treated sample is compared, but not samples from diseased vs non-diseased.... Please help, my brain is boiling... Am I at least in the right direction? thank you

#2 Trof

Trof

    Brain on a stick

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 1,198 posts
109
Excellent

Posted 14 September 2010 - 04:40 AM

Delta delta Ct can be used in any comparison, you just need to select a calibrator, "normal" or "control" sample. But this method doesn't take into account different efficiencies of your housekeeping gene and gene of interest. You can use efficiency correction method from Pfaffl.

One thing I don't really understand is that you say you normalise "theoretically", you either normalise to a defined gene/genes or not. Also the usual workflow is RNA -> reverse transcription -> cDNA -> quantification (this can be done in two or only one step), as you can't quantify RNA itself, so I assume you did this. The amount of RNA put into reverse transcription reaction is supposed to be the same. Then your housekeeping gene/genes Cts should be similar. You run reaction in duplicates or triplicates and try to get low variations in Ct within them, that's where the big stdev values later comes from.

Our country has a serious deficiency in lighthouses. I assume the main reason is that we have no sea.

I never trust anything that can't be doubted.

'Normal' is a dryer setting. - Elizabeth Moon


#3 Upa

Upa

    member

  • Active Members
  • Pip
  • 5 posts
0
Neutral

Posted 14 September 2010 - 08:01 AM

Delta delta Ct can be used in any comparison, you just need to select a calibrator, "normal" or "control" sample. But this method doesn't take into account different efficiencies of your housekeeping gene and gene of interest. You can use efficiency correction method from Pfaffl.

One thing I don't really understand is that you say you normalise "theoretically", you either normalise to a defined gene/genes or not. Also the usual workflow is RNA -> reverse transcription -> cDNA -> quantification (this can be done in two or only one step), as you can't quantify RNA itself, so I assume you did this. The amount of RNA put into reverse transcription reaction is supposed to be the same. Then your housekeeping gene/genes Cts should be similar. You run reaction in duplicates or triplicates and try to get low variations in Ct within them, that's where the big stdev values later comes from.

Trof,
thanks a lot for your post. I did do qRT PCR to RTranscribe RNA into cDNA, as you mentioned. But my problem is that my houskeeping gene's Cts are all over the place, stdev is very big. What I do is I try to profile platelets in diseased vs normal samples. I am trying to see the difference of fold change gene expression. I extract RNA, then I take same amount of RNA (say, 20 ng) in each well ( I do triplets for each sample)of 50 microL total volume. Please take a look at my excel, it may explain better. I am trying to use delta delta Ct method to 1. normalize each sample to its internal control, which is actin, and 2. to compare each sample to control. My problem is theoretically actin should be same (Ct) in all 3 samples as it is housekeping gene, but this is not what I get. My actin is very different . Can I use such actin as an internal control to calibrate or actin in not a good normalizer in this case?

#4 Trof

Trof

    Brain on a stick

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 1,198 posts
109
Excellent

Posted 14 September 2010 - 08:22 AM

Yes, housekeeping Ct values should be all near each other. We made a lots of qPCR on cells, inbred mice and such and that was usually the case. Now we are dealing with human samples and we encoutered similar difficulties, even when the RNA input is the same, there are differencies in Ct values in our housekeeping genes (GAPDH, HPRT) bigger than 2 Ct (and what's worse, those differencies are not same in each housekeeping). We are now unsure if these commonly used housekeeping genes are actually a good reference. My solution was to order six more commonly used genes to run a panel of samples and reference genes to select the best stable ones using geNorm.

By the way I miss your excel file, maybe you forgot to attach it?

Our country has a serious deficiency in lighthouses. I assume the main reason is that we have no sea.

I never trust anything that can't be doubted.

'Normal' is a dryer setting. - Elizabeth Moon


#5 Upa

Upa

    member

  • Active Members
  • Pip
  • 5 posts
0
Neutral

Posted 14 September 2010 - 08:47 AM

By the way I miss your excel file, maybe you forgot to attach it?
[/quote]
Sorry I attached a newer version of Excell which did not go through. this is a simplified version of my experiment. I already tried using GAPDH as a calibrator, and it was not good, then I thought actin b should work- and I see 2-6 Ct differences... devastating. I have run out of my human samples, no more RNA left, while trying to make PCR work with actin.
thanks for you help, appreciate it greately.

Attached Files






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

©1999-2013 Protocol Online, All rights reserved.