Software for analysis of raw real-time flouresence data - Analysis of tab delimited text flouresence v cycle data (Apr/13/2010 )

Hi all,
I am trying to use the MMQPCR method of Cawthon to analyse telomere length but my LC480 is unable to cope with anlaysing the data which has takes two flouresence readings per cycle. I know that MyiQ analyser software will analyse the data but I don't have access to it, or the software for any other real-time analysers.
In essence, I have the floursence measured at each cycle for each of my products but I need software to adjust the baseline, fit a curve and calculate ct values.
Any/all suggestions about papers, online resources or offers to try my data on your software would be greatly appriciated and anyone that can help out would recieve appropriate acknowledgement/authourship.
Looking forward to hearing from you.

Hi I am not familiar with your procedure (MMQPCR method of Cawthon) but I used to play with raw fluorescence qPCR data in R (http://cran.r-project.org/). I was able to re-create amplification plots, set precarious baselines (based on means and SD of pre-selected baseline cycles) and estimate Cq from a non-linear regression fit (imperfect) or from linear regression of the log-linear amplification phase (much better than nls but still experimental).

Maybe you can write an script in R to do what you want. I was surprised of how simple the script is to do these most basic things I tried. I can give you my script (imperfect) if you want. I am not good at writing scripts but if you have someone proficient in R it will be very easy for him/her to write a script to do what you want. I guess you might even do it in SAS or some programming language with linear/non-linear (depending on your choice) regression capabilities, I just suggested R because it's what I use.

Let me know if you want to take a look at my code.

Good luck!

I would do this.

1. Export raw data from the Experiment-Data tab (preferably in the HTML format).

2. Import to Excel, now you have a row for each aquisition. Use Excel script or macro (or do it manualy) to select every even row (every second aquisition) and copy to a new location. That way you separate the two product data.

3. Modify the data to fit PCR Miner format as showed in the form and submit.

PCR Miner computes Ct values from the second derivative maximum (same as LC480 does) without the need for threshold. You get your Cts and can calculate the quantification using standard functions in Excel or elsewhere.

Thank you so much for the suggestions.
PCR miner is exactly the piece of software that I needed. The online function is great and I have applied for an executable software version.
Best of all it has confirmed that my experiments are working
I do have one followup question about baseline subtraction; I have been using the mean flouresence over the first 15 cycles, does anyone have a better suggestion or any critisim of this approach?
Thanks again.

I guess if none of your reactions produces significant amplification before cycle 15, it is safe.

I once had one sample that amplified earlier than 15 (Cq was about between 10 and 11), it was an unknown sample and this was completely unexpected. I had to extend my linear dynamic range (another new set of dilutions) to include this sample and specify a baseline at an earlier cycle. This messed a bit my other curves, so from then on I started to adjust the baseline for each reaction individually. It has never happened again (it was only those sample from one specific treatment that gave this response), so I guess it should not be really common (at least not in my experiments) that you get significant amplification very early. The good thing is that if your baseline definition is not good it will most likely be very obvious in the plot.

bison on Apr 13 2010, 05:41 AM said:

bison on Apr 13 2010, 05:41 AM said:

Hi Bison,

The following profile should work on the LC480:

1. 95 degrees x 15 minutes
-----------------------------------

2. 94 degrees x 15 seconds

3. 49 degrees x 60 seconds
Repeat steps 2 and 3, one more time
-----------------------------------

4. 94 degrees x 15 seconds

5. 59 degrees x 30 seconds
Repeat steps 4 and 5, 3 more times
----------------------------------
6. 85 degrees x 15 seconds
7. 59 degrees x 30 seconds, with signal acquisition
Repeat steps 6 and 7, 19 more times
----------------------------------
8. 94 degrees x 15 seconds
9. 84 degrees x 10 seconds
10. 85 degrees x 15 seconds, with signal acquisition
Repeat steps 8, 9, 10, 26 more times
----------------------------------
Melt program: 59 - 95 degrees; 0.5 degrees, 5 seconds, per step

Please note: during the cycling of steps 6 and 7, the telomere product is exponentially amplified, while the albumin gene product remains double-stranded and unavailable to its primers. Also, during the cycling of steps 8, 9, and 10 the albumin gene product is exponentially amplified, while the telomere product remains completely melted and unable to hybridize to its primers.

Primer sequences and concentrations:

telg and telc primers: sequences as in the attached paper, final concentration 700 nM each.

albumin primers (I recommend these over the albumin primers that are in the paper):
albugcr1: CggcggcgggcggcgcgggctgggcggAAAcGCTGCgCAGAATCCTTG, at final concentration of 200 nM;
albdgcr1: GcccggcccgccgcgcccgtcccgccgCTGAAAAGtAcGGTCGCCTG, at final concentration of 200 nM.

Serial dilution for standard curve: five concentrations, three-fold dilutions, with the highest concentration being 60 ng of DNA per 10 microliter reaction.

Master mix composition as in my 2009 NAR paper.

I observed no "No Template Control" amplification using the above protocol.

Richard Cawthon

Hi, Bison:
I am using the LightCycler 2. And I am confused with the data I exported to the excel file. So how did you treat "two flouresence readings per cycle"? Do you use excel script to translate the data for the PCRMiner or use Rscript?
Thanks for your help:)