How To Perform The Delta-Delta Ct Method (In Excel)

[@StevenBradburn]

THE ONLINE GUIDE toptipbio.com/delta-delta-ct-pcr/

[@patrickgonzalez3303]

Hi! I watch your video 5 years ago when I was learning how to make a qPCR analysis. Five years later, I am currently doing a PhD and just needed a refresh on this topic before doing an analysis. Let me tell you that your video is still precise and easy to follow! thanks!

[@abdulrazakbaha5070]

Thank you for the insightful video! Your explanations made learning enjoyable and much clearer!

[@asamoahbenjamin8164]

this is an excellent tutorial. precise and concise.

[@danielleonardoalzamoraterr9472]

I have had a little problem understanding that, but your video has really help me now! Thank you a lot.

[@panizshirmast1114]

very useful! i wish u were adding these final numbers to a bar chart and show how to make it.

[@deon8252]

Simplified, short and precise..thanks for this great video 👌

[@StevenBradburn]

Deon Kumbirai Mandebere thank you :)

[@ellengarcia3467]

Thank you! You made it so clear and easy to understand.

[@biohazard6368]

1. If I normalize the results (control as 1), should the graph show the mean and standard deviation for these normalized results? 2. Which results should I take for statistical analysis? Results after normalization or any other, e.g. log 2 ^ -ddCt? If there is a significance for results after log, can I mark it on the graph with results after normalization? Thanks in advance :)

[@TheEduardoortega]

How would you go about calculating the error for 2^ΔΔCт?

[@zianasif9247]

Amazing video !! Thanks Steven

[@hannaschneewei4612]

I will thank you specifically when i get my diploma.

[@anjalikrishnaa6940]

You have explained this so well! thanks a lot!

[@AhmedKMahdi-um3yg]

What about a calibrator? You calculated the DDct between 2 groups (control & treated) But what about having one sample (calibrator) that is usually used to calculate the fold changeby normalising everything to it?

[@namrithamaniyodath2877]

here the calibrator is control. so in this video delta delta Ct value is calculated by subtracting the average calibrator/control delta ct value from the individual delta ct value. I am 4 years late, i hope it helps someone!

[@amchi6688]

Thank you very much. You have explained it very clearly.

[@MrFilu13]

great teaching

[@shrouqalmaghrabi7383]

Great explanation ... Would you explain the error bars, please?

[@Yousha999]

Thank You so very very much for the clear understood explanation

[@Sid-ro2ye]

Hi! Great video, just wondering on how you would plot your data? (fold change) Would you take the log base 10 of the fold change to plot and use the standard deviation (error bars) of the 'normal' fold change or the log base 10 fold change fo SD? Many Thanks!

[@eriklund9747]

Cutler Beckett tired of chasing pirates and naturally got into genetics and excel tutorials

[@panizshirmast1114]

how can i add SD and error bars in this results? as we only have one amount for delta delta ct

[@DrRachie]

So helpful. Cheers.

[@nomizeeshan3031]

If we have different tissues values of only control than?

[@willians.bernardes7478]

Great. Thank you!

[@nabinghimire2006]

what to do if we have two genotype with control and treatment in each group?

[@Idonai]

Why can you take the average of the fold changes? A fold change is generally asymmetrical between 0 and infinity. Shouldn't you take the average of ddCT and then calculate the fold change? e.g. if you had the 2^-ddCT values 0.5, 1, 1.5 the average would be 1 even though 0.5 and 1.5 are not relatively the same distance from 1.

[@judith4294]

Thank ypu so much! it was so helpful your video. I have one question, what happens when the gene of interest is more expressed that the housekeeping gene?, how can perform the calculation? Thanks in advance

[@amenehghaffarinia5642]

Many thanks for this great video. Please add more related videos.

[@StevenBradburn]

Thanks! I am in the process of making more :)

[@arshadbhakkar]

Very precise and understable thanks

[@austinwhitman2285]

Why do you average the delta-CT values of controls 1, 2, and 3 together? Are controls 1, 2, and 3 from the same group, and if so why are they treated as separate when averaging the CT values? The only way this makes sense is if each condition was an n of 6, where controls 1,2, and 3 are all the same cell line or whatever. The labeling is not clear.

[@monicacitos]

ThaNk You so much

[@mebratuify]

Thanks, but why why do have two Ct values (Ct1 and Ct2)?

[@khairunnisafitri138]

understandable, thank yiu so much sir

[@torlarsen2212]

You are great!!

[@omarbabena8199]

I always get a value a bit higher than 1 when I averagy control group? Something like 1.16, it should be exactly 1 right?

[@paveluvarov7765]

Try to calculate geometric average

[@adrianoluis4328]

Great tutorial! Thanks!

[@StevenBradburn]

Thanks Adriano

[@galaxyeyes1457]

Thank you SO MUCH!!! 😭

[@StevenBradburn]

You're welcome!!

[@jyotibakshi4681]

sir can we use housekeeping gene as positive control

[@PetoBewise]

Hi, I've got a question... so my results are average control fold change = 1,6 and the group of interest average change fold is 0,7... Now, I dont quite understand what the fold means. My control is 1,6 fold to... what? Or should I devide 1,6 with 0,7 and then I'll get the fold between them? Thank you very much.

[@StevenBradburn]

Hi BeWise So essentially your group of interest has a lower gene expression than your control group. What would be better to do now is to normalise the gene expression levels to the control group average. This way, the control group average value will be '1' and the group of interest will be relative to this group. So, make a new column and divide all of the gene expression values (2^-DDCt) for all the samples by the control group average 2^-DDCt (which is 1.7). Then average these values for the controls and the treated. The control average should now be set to '1'. Does that make sense? Thanks for the comment!

[@PetoBewise]

Yes, it does. Thank you very much!

[@StevenBradburn]

BeWise excellent! Very welcome :)

[@unknown9106]

BeWise cf

[@shilpasharma4383]

Very helpful..thank you sir

[@christianhinze1701]

Thanks for the video. However, usually I want my control 2^-ddCt values to have mean 1 in the end. I do not get this all the time by using your approach. Any comments on this?

[@StevenBradburn]

To normalise the gene expression levels to the control group average so the control group average value will be '1', make a new column and divide all of the gene expression values (2^-DDCt) for all the samples by the control group average 2^-DDCt. Then average these values for the groups and the control average should now be set to '1'. Does that make sense?

[@christianhinze1701]

Absolutely. It is just never mentioned every time the method is presented. I find that odd.

[@StevenBradburn]

@@christianhinze1701 sorry about that. It is something that I realised after I made the video. Hopefully I can update this to incorporate this point

[@Knickerbocker__]

@@StevenBradburn I don´t quite understand the last part "Then average these values for the groups and the control average should now be set to '1' ". So if i do this i only get one value for all the controls together and all the treated together, right?

[@doctorzoology1545]

@@StevenBradburn i have same question...if this is the case then there will be no SD or SEM for control group? is it possible to plot controls without any SD?

[@chloev9275]

what if your first Δct is negative?

[@ayeshashaikh6678]

This was great! Thanks.

[@StevenBradburn]

You are welcome :)

[@borainhav]

Thank you so much! I found this very useful indeed and nicely didactic. Do you know if this relative quantification method can be used to compare transcript levels between a mutant and its corresponding wild type background? In my case, my mutants are T-DNA insertional mutants that are expected to interfere with the production of the transcript. Also, I guess one could apply the method even if having only one biological replicate? I understand it would be much less accurate but still valid? In that case I would have only one value of ddCt for both groups and in the case of the control group, that value would be cero.

[@pavithrarv9997]

Thanks for a great video

[@sitihamidatulaliyah9587]

thanks a lot

[@madhavdhakal1715]

Very helpful

[@탁-z5l]

Thanks a lot. Very simple and easy. ^-^

[@RajeshKumar-xm9pc]

could you also post a video on differential gene expression using qpcr data sets and volcano plor

[@StevenBradburn]

Sure, I will add it to my list :). There is a volcano plot video coming out soon using GraphPad Prism if that helps?

[@Phytopathology]

Oh man! I owe you

[@anjalimohan7570]

nice lecture!!!!!!!11

[@dinaharys2241]

Hi, i've got a question, What does the fold means, for example if the 2^-delta delta CT value for your treated group is 1.7, does this mean that the expression in your treated group is 1.7 higher compared to the expression of your housekeeping gene, or compared to the expression of your untreated group?

[@BlackestSilver]

1.7 higher compared to the expression of your housekeeping gene

[@cesaresgua]

Excellent tutorial! What do we do in the case when we have multiple targets amplified under the same conditions? Thanks!

[@StevenBradburn]

Thanks Cesar. So, you will need to repeat the analysis process for each target gene you are interested in. All the best, Steven

[@cesaresgua]

Thank you! that makes sense. Each target have to be compared with its respective control.

[@Lineee151]

Is it normal that my average for control in the end doesn't come out as 1?

[@StevenBradburn]

If you want to normalise the gene expression levels to the control group average, so the control group average value will be '1' and the group of interest will be relative to this group, then: 1. Make a new column and divide all of the gene expression values (2^-DDCt) for all the samples by the control group average 2^-DDCt 2. Then average these values for the controls and the treated. 3. The control average should now be set to '1'.

[@KitanonnonN]

@@StevenBradburn This comment save my life

[@doctorzoology1545]

@@KitanonnonN in this case there will be no SD for control group...but in many articles SD for controls has been shown. Is it appropriate to show control fold change without SD or SEM?

[@ifeanyichukwueke9942]

Nice video!

[@ladushky1]

Thank you very much for the great explanation. Do we raise delta-delta Ct value as an exponent of 2 due to the diploid nature of the genes? If we were looking at yeast for example, would this still work?

[@StevenBradburn]

Thanks! So the exponent of 2 is used because it is assumed that for every PCR cycle, the amount of product doubles (i.e. the Delta-Delta Ct method assumes the PCR efficiencies for the gene of interest and reference/housekeeping gene are 100%). Sure, this should work for relative gene expressions in yeast. Other relative methods, such as the Pfaffl method, can adjust the exponent of gene of interest and reference gene separately to account for differences in efficiencies. Hope that helps, Steven

[@Masanelesz]

@@StevenBradburn if both pairs of primers have 94% efficiency, is it correct to perform this method by changing the power base by 2*0.94 = 1.88?

[@Athams]

I want to ask, how do you present this data? Bar charts, or histograms?

[@StevenBradburn]

It would be best to present as a scatter plot to show the spread of data in each group; however, bar charts are the most common way of presenting gene expression data

[@nunezdp]

i liked the video simple and clear. Can you add the reference to cite this in my document???

[@StevenBradburn]

Hi Diego, Many thanks. Here is the Delta-Delta Ct method paper: www.ncbi.nlm.nih.gov/pubmed/11846609 Thanks, Steven

[@worldofinformation815]

Thank you Sir

[@KhaLed-pb4pu]

Is this an INCREASE? OR Decrease? can we have negative values? Fold?

[@m1stylark]

i want to know too

[@melisaelianamagallanes6843]

How shoud I do if I have two hkg?

[@TheNavigator]

nicely presented!

[@gp235801]

Is it necessary to take the average of control group's delta ct values? I believe the Livak method doesn't include to make this average. You can simply get your delta delta Ct values by subtracting the delta Ct of control from delta Ct of treated samples. Please explain.

[@StevenBradburn]

It's not a requirement and really depends on the design of your experiment. If you have a cell culture experiment, for example, then it is easy to use the control from that experiment. However, if you are performing qPCR on 2 groups of human biopsy samples, then there is no specific control sample per say, hence why the average (or geometric average) will come in use. Alternatively, you could just select 1 sample from the control group to act as the 'calibrator' sample. Hopefully that made sense

[@gp235801]

@@StevenBradburn Got it! Thanks so much.

[@shamimaislam1103]

Hi, i have two different samples with same conditions can i use delta delta Ct method for quantifying them

[@StevenBradburn]

Hello Shamima, Technically yes. But the more biological samples you have the better. By having 2 samples only, 1 sample will have a delta delta value of 1 and the other will be relative to this. Also, it wouldn't make much sense to compare 2 samples with the same biological conditions. Ideally, they should be a control and treated group, for example, for the results to be informative. It may be worth looking at the delta Ct method (2^DCt) Many thanks, Steven

[@shamimaislam1103]

Thank you very much for your answer, I have more biological samples , but my concern is to compare within these two groups (samples) and I don’t have control as delta delta ct method suggests to see relative gene expression from reference sample(control or untreated)

[@shamimaislam1103]

Top Tip Bio I m sorry to bothering you do you have excel file of 2^dct , could you please share

[@StevenBradburn]

@@shamimaislam1103 i am currently away at the minute. But when i get back i can do yes :)

[@Mariiamdov]

another question - some of my samples have N/A for the target gene because they have not been treated with virus so should not any viral gene expression. Can I follow this method exactly or is there another way to account for Ct value of 0?

[@StevenBradburn]

Do you need to analyse this data? If there is no detection (ie, Ct value) in controls and there is in treated samples, there is no need to perform a statistical test on this data. The results are clear to see

[@cristinacervantes713]

What if the difference between CTgen - CThousekeeping is negative? Is the delta ct always positive?

[@StevenBradburn]

Hi Cristina If you have a negative dCT this indicates that your gene of interest is expressed more than your housekeeping gene. Do you anticipate that your gene of interest is highly expressed? Have you confirmed the qPCR reaction is efficient and does not contain primer-dimer? Thanks, Steven

[@Mariiamdov]

Hey thanks so much for this clear video! I have two questions. I have a treated group who received infection and a mock group who did not. Are they the equivalent of your control group? Secondly, how would you plot this information on a graph to display in a report? I am a super novice at qPCR!

[@StevenBradburn]

Hi, So your groups sound equivalent to the example (mock = control and infection = treated). Usually people use bar graphs to plot the mean and standard deviation. I recommend plotting all of the data points instead of a bar graph. It all relates to the distribution of your data. But, if you have only a few samples, then plot each point instead of a bar graph.

[@MA-yt4xf]

​@@StevenBradburn Hi, I have the same amount of samples as yours in the video and I have been asked to plot it as a bar chart. So I should do Avg fold change of treatment group divided by Avg fold of control group, correct? Then plot that value as the height of the bar graph. Also, for the error bars, which values should I take into account and how? Thank you!

[@mehrdadrabiei5811]

Is delta-delta Ct used only for relative qPCR? May I use delta-delta Ct for analyzing absolute qPCR data? Please answer me soon. Thanks

[@StevenBradburn]

Hi Mehrdad The delta delta ct method is only for calculating relative gene expression values. For absolute gene expression analysis, this requires a standard curve. I am in the process of writing up an article for our website on how to do this. I will provide this soon enough when it is finished. Best wishes, Steven

[@mehrdadrabiei5811]

@@StevenBradburnCould you please tell me know how can I analyze the data for the absolute qPCR? Copy number for standard curve points provided and after running qPCR assay I got copy numbers for unknown samples. I need just hint what should I analyse mathematically. I did not have a reference gene. I had the unknown treated sample and the unknown untreated sample as a control. That's all.

[@StevenBradburn]

@@mehrdadrabiei5811 Hi Mehrdad. This PDF should help from Illumina. Refer to step 5. www.illumina.com/Documents/products/technotes/technote_eco_absolute_quantification_using_sybrgreen.pdf

[@kania-hiro836]

How to make the error barr in this graph?

[@StevenBradburn]

Hi Kania-Hiro, Does this tutorial help? kzbin.info/www/bejne/bHPGlJaKfcRlf8k Thanks Steven

[@vanessaleikooikawacardoso6949]

Nice video! Helped me a lot! :) But I have one question: how do I calculate delta-delta Ct using 2 housekeeping genes? I'm using 18S and beta-actin genes.

[@melaniee467]

You calculate the geometric mean of the CT value for these two reference genes, for each sample you have, then you use this geometric mean as the control in this video

[@enasmohd2508]

Thanks

[@leticiaolivera-castillo8638]

very nice

[@KhaLed-pb4pu]

So what does the result tell? That in the treated group, the gene of interest was expressed 54.9 fold less? or more?

[@lisaengel4128]

It’s 54.9 fold the control - meaning higher gene expression. You can kind of already see it by looking at the Ct values. Lower Ct means more template, which means higher gene expression (if cell number was the same). The reference gene is to normalise any differences in cell number.

[@KhaLed-pb4pu]

@@lisaengel4128 fold the control is not English. fold MORE or LESS than the control is English. I can see Ct values, and if it was clear, I would not ask the question and why bother even do the whole calculations?? My question fold increase or fold decrease.

[@lisaengel4128]

kwi01 kwi01 well, I think the problem is not the English language or the Ct value, it’s clearly your lack of understanding the mathematical part. If it was less it would show some value less 1...

[@dinahkarimi6969]

I Using two ref genes...how do i go about

[@bennie1450]

Did you find a solution? I have the same issue.

[@thaisandrade9783]

Me too ..

[@huitinglin6277]

The last steps calculate 2^-(delta-delta Ct), why the formula power "negative" delta- delta Ct? what does "the power of negative" mean?

[@wintersyixx]

Was thinking maybe they want to remove the negative value and to accommodate logical explanation for fold change in whole number, instead of a negative value. that's my guess.

[@fazeela1548]

I don't understand the treated 1,2,3 values from where we get these?

[@StevenBradburn]

Hi Fazila, The Treated 1,2,3 samples represent different biological samples in the 'Treated' group. So these can be 3 repeats of the same experiment. I hope that helps. Thanks, Steven

[@fazeela1548]

@@StevenBradburn thank you so much I got it👌😍

[@zeeshanqasim8925]

hello sir, how I can solve this equation Gene copy number= 10^(31 Ct)/(10/3)? Thanks

[@liopleurodon155]

31÷(10÷3) is equal to 31 × 3÷10 which is equal to 31 × 0.3 = 10.3 (~ 10) Your gene copy number is roughly 10^10.

[@Gooch_cruiser]

Not sure why you average the control delta ct. Each biological replicate should be compared to itself and not the average of all the control experiments. If you just subtract the control delta ct of individual experiments from itself (13.38-13.38) then its more honest and you get the same result. Because 2^0 is always 1.

[@mehrdadrabiei5811]

Hi, how can I analyze my data for absolute qPCR? thnx

[@Utada379]

Isn't it a technical replicate? Not a biological one.

[@TheBleachedazn]

It's possible that these are cellular replicates (pseudoreplicates) and not true biological replicates you would have if you had three different animals? I'm also not sure why he would average the control delta Ct but not also the treated delta Ct to remain consistent.

[@fatihaislam9149]

Yeah I was confused when he was working out DDCT like that because my data was calculated in the way you’re explaining

[@N5033]

Very good son

[@liopleurodon155]

But that 55fold in the end means lower not higher, right? The Ct values for the gene of interest are way higher which means less copies compared to the housekeeper.

[@StevenBradburn]

Hi, So, the analysis does not compare the gene of interest to the housekeeping gene. Instead, the housekeeping gene is used to standardise the result. Instead, the analysis reports the difference in the gene of interest expression relative to a comparison group/sample. I hope that helps, Steven

[@Ice84letters]

Why did you average the control group? then the value will be zero with a fold change of 1, which is the value to compare the treatment samples, why you should have a standard deviation from your control groups in the experiment? i do not really understand this, in my lab they do it the same way but i have read a lot and there are people who is against this... the fold change should be 1 and the log10 there should not be standard deviation, could somebody help me with this doubt?

[@StevenBradburn]

Hi Irene, Many thanks for your comment. I used the average control delta Ct since this will enable the calculation of 2^-(∆∆Ct) for all the samples, including the individual control samples. Other people just match the experimental samples and determine the relative gene expression ratios separately. This is all well and true for experiments that have matched pairs, however, this is difficult when the two experimental groups vary in n numbers and do not have matched pairs. Another way to select a calibrator/reference sample is to pick the sample with the highest Ct value, so the sample with the lowest gene expression. This way, all the results will be relative to this sample. If you want to get an overall average fold change of 1 for the control group, you can normailse the results. To do this you would make a new column and divide all of the gene expression values (2^-DDCt) for all the samples by the control group average 2^-DDCt. Then average these values for the controls and the treated. The control average should now be set to '1'. I hope that makes sense? Thanks!

[@Ice84letters]

Top Tip Bio my last question, control1, control2,etc are biological replicates and the ct1, ct2 technical replicates? Thanks for the answers!!

[@StevenBradburn]

Irene Cartas Hi Irene. Yes that is correct. Sorry if this was misleading. Hope it helps you :)

[@Ice84letters]

thanks a lot for your answers!

[@Ice84letters]

smotra.site/video/n3SZlsoKOoo/4-5-cuantificaci-n-de-la-expresi-n-g-nica-me-todo-del-ct.html i am surprise to find this other way...in general i see in both cases the same behaviour but the fold change...is not exactly de same, so finally, i do not really know how to do it, and now i am reading also there is another way...using the formula publish by PfaffI

[@erichmullerrivera]

SUPERB!

[@NovaBio]

Would you like to collaborate for educational video?

[@onderozbasak]

I am more imperessed by how fast you could write Δ as if it is normal key on the keyboard, rather than your whole explanation of this technique.

[@StevenBradburn]

Copy and paste is my best friend

[@mikevankemenade6379]

Control+shift + q switches to symbols in word. Your keyboard becomes greek for one symbol. So for Δ control + shift + q and then shift plus d (because its a capital d). And boom a Δ

[@emilkaykov9892]

in the "average Delta CT", may I ask why you only take the first 3 and not the whole column?

[@StevenBradburn]

Hi Emil, The results of delta delta Ct are relative to a control/untreated sample (also known as a calibrator). Therefore, this is why I only take the average of the 'Control' group, instead of the whole column including the 'Treated' samples too. I hope that makes sense. :)

[@emilkaykov9892]

It does ^^ thanks, great video btw. Just kind of a last question, my "2^-DDcT" results are in very large number ex: 600, 3000 and 10000 is there any way to normalize this or just take the data as it is.

[@StevenBradburn]

Thanks, really appreciate it. Wow, sounds like a seriously strong change in gene expression. You can normalise the results to the Control group average 2^-DDCt, if you have a control group? This will basically make the Control group 2^-DDCt equal to 1. But I doubt this will drastically change your results because you have such big numbers.

[@emilkaykov9892]

Nevermind I realized there was an error on my behalf, I got all of my numbers between 0.5-2, besides one which was 600 :D (weird). Anyway, thank you again for taking your time to answer!

[@TheEduardoortega]

How would you go about normalizing to the control?

[@akarsoo5338]

Hello. Please correct your instructions. In the ddct of the treatment groups, the dct should be subtracted from the average dct of the treatment, not the average of the control.

[@AmjadKhan-fd5nx]

Please share your email. . Thank you

[@MukwayaVincent]

Many thanks for a very precise video

[@MundodePolly]

Thanks