Dear Vivian,

This looks alarming indeed.

Unfortunately, I do not use the LightCycler 480 and therefore do not know if this is something due to eg very low fluorescence levels. Do these levels [fluorescence max 11] correspond with levels you got before for nice smooth curves? Because if you only included your test samples without a positive control that performed fine in previous experiments, the fluorescence scale might have been stretched by the software, but after all negative if compared to a real positive sample.

Anieta

Dear Vivian,

I use lightcycler, but 2.0 version (is it different from 480 ?), and not in a quantitative manner, but just for detection of SNP.

However, when I see your curves, they seems to me a little strange compared to mines (I join you the short guide of my machine, where they give samples, that I have).

Indeed, and for me , fluorsecence increases with amplification, till about 30 cyccles (yours conitnue up to 40 cyles), then reaches a steady state; that is not your case (straight lines). What is the mean of calculation of your machine ? (end point, at midle curve ?) and does it necessitate to have a steady state to be able to give you a result ? (which is not your case).

You don't give us, but what are the corresponding melting curve that allow to have a peak by log transformation corresponding to your awaiting result (see my short guide).

But peharps all I describe to you doesn't correspond in any way to your awaiting curves and results.

Best regards

Didier J

Dear Vivian,

You need to follow Anieta's suggestion and run some samples you know works to eliminate mechanical problems and that should be your first step. Looking at your attachment I don't think your samples have a S-shape it's more like a linear curve indicating low PCR efficiency. Let us know how you go.

Cheers,

Claus

Hello Vivian,

Have you checked if melting curves for the primers pair you are using looks ok?

Which efficiency you have on those primers?

Does not seem to me a technical problem (such lamp alignment)... I would think to a kind of inhibition happening during the amplification either due to chemicals or excessive amount of template.

Leonardo

add to all, i should say the amount of cDNA u put in is also important. i have seen ,when cDNa has been used in a big amount in real time mix,the reaction will be inhibited with the reagent which exist in the cDNA mix.so the reaction will not exceed.

dear Vivian,

from your TIFF file one can see that you are in the linear region of RFU vs. cycle (or time), so first derivative approach to find out Cp value is not well suited (this is the first default choice when you pull down analysis options). The second one, would give you Cp values even for green curves...Now, Tm PCR program will melt your PCR product (separate DNA strands) and monitor change of RFU vs. temperature. The temperature when RFU change is 50% of maximal ---is Tm. This curve has different resolution from amplification curve...and all you need is that you PCR was working (you have product)...so info about increase in RFU vs. cycle (your shacking curve) is not influencing...

Hope this helps...nice work in any case...best regards, ivan brukner

([email protected])

Dear Vivian,

Are you using a commercial kit ready to go? What the amount of mRNA do you use to make cDNA?

do you check the quality of mRNA? How? in Agilent system:? or by regular gel eletrophoresis?

What the concentration of primers? and amount of cDNA used?

the amplification look like iniciating at 20 cycles , what about your internal control like, GAPDH, e others? How did you choose and where are the curve of this gene?

I advice you to start with only the internal control and a positive and negative control with a very acurated concentration of mRNA /cDNA , after this experiments , you can conclude if your reagents is in a ideal concentration or are workin in your assay.]

Have a good look, if you want more discussion. just write me.

Mario

Please follow the procedures and correct protocols to get good results .

Gud luck

Dear Vivian,

It ggenerally happens when the standard curve exceeds the linear range of detection. Have you tried to dilute your template before using it for real time PCR amplification ?

Adriana

Try different DNA concentration

Use new working reagents and increase template quantity, You will definately get good results.

Wish you all the best

hi, i used Light Cycler 480 before, even i had the same result sometimes. COuld you check your dilution factor/concerntration of cDNA and primers.

Vivian,

I would agree with most of the comments before. Trying different DNA concentrations should solve your problem.

Good luck

Roland

I agree with Tim, your PCR did not work at all. your curves don't reach a plateau phase. I do agree to most of the trouble shooting suggestions here, to make the PCR work.

But most important I want to mention, that the Light Cycler 480 is not at all ok with all seals on the market. I had a similar result once, when I did not use the ROCHE film and also plate. While using different plates generally only results in lower fluorescent signals, using different seals (for example the BioRad seal for myIQ, which has a lot of glue on it) would cause such staggered curves. Make sure you are using compatible seals and plates. I have tested several plates and seals for our lab and am happy to share results, if you're interested.

Best,

Nora

The picture of the Amplication curves you have got there are linear view.

1.Did you have a look at the log part of the curve?

2.Did you compare your negative controls with that of your samples in the log phase? Are they coninciding together at the same point?

3.It can be clearly seen that the other samples shown in green might also be positive if the RED line is your positive control.

4.Retry the experiment if you have sufficient sample(already confirmed positive) left along with positive control,negative control and one sample to be re-tested. Check if you are getting similar results.

It could a problem of the calibration of the machine which is why at times you get jagged lines. There could be other reasons such as your probe might have got quenched (if left too long to expose in the light).

Are other experiements run in that machine showing similar results at the end of the run ? If it is then it is definitely machine calibration problem. If it is not it the problem with your experiment in which case you might want to try and alter by not exposing the probe to light too much !

Good luck !!

Thank you very much for everyone's reply.

I must add that each time I do this experiment, I set up 3 distinct amplification efficiency curves (0-60ng MCF-7 cDNA) with each curve/set added with different concentration of MgCl2 (3mM, 3.5mM, 4mM), so to optimise the MgCl2 concentration for future experimental studies. The previous amplification curve image I've uploaded is the one with 4mM MgCl2. The 2nd image I uploaded with this reply is the set with 3mM MgCl2 which was on the same microwell plate, same run, same type of reagents. As you can see, the curve from 3mM MgCl2 set look more ideal but fluorescence was so high (65 units), which could have stretched the fluorescence scale (as suggested by Anieta).

The other thing is, when setting up the 4mM MgCl2 set, I ran out of the PCR Master Mix stock I have been using and opened a new stock/tube, which makes me suspect the integrity of the master mix I just opened. I am currently running a standard PCR to check if the mix is working well.

I have repeated the efficiency curve experiments using a new plate with the same settings, same reagents and I have attached the resulting amplification curve with this reply. The curves look flat and linear with low amplification. The % efficiency was 150% which isn't within the 95-105% ideal range. The corresponding melting curve (attached in following reply) also looks shifted.

Melting curves

Vivian,

your amplification looks better than the first time and the melt curves indicate that you amplified something. But of course your curves still look odd. If you look close to the amplification curves, you will see, that you are only for a short period of time in the exponential phase of amplification and your reaction then seems to stop and signals go up. I have attached some of the tests that I ran for our lab using different seals, plates and dyes and you will see that the curves look in some cases similar to yours. Therefore, I again want to ask, which seals and plates and also dyes you are using for your reactions. To me it very much looks like something goes wrong with the seal or plate (or both) and not with the reaction itself. The efficiences of 150% are not at all reliable.

Best,

Nora

Dear Vivian

Hi

Did you used the same thresould value in all of your experiment?

If you can, please send me the pic. of the curve. I could not download it. Sent it to my e-mail: [email protected]

ALso, did you run a reaction with genomic DNA to check the health of the reaction?

To begin with, the curves you attached look nothing like PCR reactions. So my first and foremost suggestion would be to revise your experiment. Redesign your primers and check that your target sequence is not in a low complexity cluster.

Moreover, whatever your final experimental setting will be, you cannot do this without a positive control. Had you put a positive control in the original experiment, you would have a) seen the difference of shape between a sigmoid curve and your amplification, and b) your plots would have been in a proper scale.

You might check with RepeatMasker or any other similar tool.

http://www.repeatmasker.org/webrepeatmaskerhelp.html

The first set is clearly a negative reaction. And though the LightCycler actually shows you something, and tries to calculate what it can, it is all bogus data. The jaggedness of the plot indicates that the scale is zoomed in to its max. Put beside a positive control, you basically would see a flat line.

IMHO, the second set of reactions comfort my idea that these are more artefactual than actual PCR target amplifications, in particular because:

1/ diluted template curves behave chiefly as the concentrated ones, but the slope changes.

Thus, to me it seems unlikely that the issue is a lack of reagents or enzymes.

2/ the curve is not a typical sigmoid. It is a more linear reaction, like a runoff from a repeat element replicating.

The first thing I would recommend is to run an agarose gel to check the actual size of the amplification product against the expected amplicon size.

Then, it is a matter of cost-efficiency:

a/ the current PCR setting is not critical (ie first setting of the experiment), in which case I suggest you redesign the primers, with particular attention to repeating elements in the gene. You should at the very least BLAST your primers against the species genome.

Your cDNA might also be at fault. Obtaining cDNA from an third party that has a functioning PCR system would be a very good positive control.

b/ the current PCR is set.

Optimisation of the reaction must be done.

First, doublecheck that your amplification product is what you think it is.

Check:

Size on an agarose gel.

Absence or limited amount of secondary amplification products or primer dimers

If necessary, sequence the amplification product

Once the basics are covered:

You should prepare a fresh batch of PCR reagents: fresh dNTPs solution, Taq buffer, etc.

An important parameter you could use for optimization is MgCl2 concentration, since Mg++ ions are rapidly depleted in excess of dNTPs.

Another useful reagent for optimization is DMSO.

And, of course, the annealing temperature.

As before, your cDNA might also be at fault. Obtaining cDNA from an third party that has a functioning PCR system would be a very good positive control.

But, unless you really need to do this, I would strongly recommend experimental redesigning first.

In your particular case, though, optimization should definitely not be your immediate concern.