Hello everyone,
I am extremely confused about analysis of qPCR data after virus infection or stimulations, and I need help. I hope I can explain properly.
Let's say we have two different cell lines.
1. HFF WT and
2. HFF protein A knockout.
We infect (or stimulate) these cells with virus and then do IFNB1 and GAPDH qPCR.
So, we will have these conditions:
- Mock HFF WT
- Virus-infected HFF WT
- Mock HFF protein A KO
- Virus-infected protein A KO
Let me try to explain what I understand how people analyse this kind of data:
First, you get the Cq values of IFNB1 and GAPDH.
Then, you do (IFNB1 - GAPDH)
Using (IFNB1-GAPDH) values, then you do (Virus infected HFF WT- Mock HFF WT) and (Virus infected HFF protein A KO - Mock HFF protein A KO).
Then you are doing POWER(2; -(Virus infected HFF WT- Mock HFF WT))
or POWER(2; -(Virus infected HFF protein A KO - Mock HFF protein A KO).
My problem is:
Since the mock cells are not stimulated, they usually give either negative or very high Cq value of IFNB1. (for example: 36, 38, 39). And this value is totally randomly changes experiment to experiment. Sometimes Mock HFF WT is 36, and the Mock HFF protein A KO negative. Sometimes vice versa. These also changes between biological replicates all the time. (I usually write 40 when it is negative, which is my cycle number).
In this case, can I just consider all the mock cells will have negative value and consider their Cq is 40 (eventhough I get 35-39). Of course they can have some IFNB1 even they are not stimulated/infected and this can change between these two cell lines, but having different mock results changes the whole outcome.
Note:
My IFNB1 Cq values are usually 20-26 in infected cells.
My GAPDH Cq values are usually 20-25.
I use GoTaq.
I convert total of 400 ng of RNA to cDNA (in 20 uL).
Then I take 1 uL for qPCR, so 20 ng. Can increasing the cDNA amount help me to get positive values from Mocks and equalize them?
Thank you very much.
"Then I take 1 uL for qPCR, so 20 ng. Can increasing the cDNA amount help me to get positive values from Mocks and equalize them?"
Ct values are mostly determined by template abundance if the reaction is optimized (after primer specificity & optimal mix). If you get 25 Ct with 400ng of RNA, then if you input 800ng then you will get ~24 Ct because PCR is a doubling process. I encourage you to try this resources permitting: find a sample with ~20 Ct and do half dilutions with nuclease free water. You will find that each dilution will push the Ct by roughly 1 cycle. Or if you did 1:10 dilutons as people do with standards, 3.3 cycles. If you are using the delta delta Ct method for normalization, you shouldn't be trying to make certain samples go up or down in expression to make them match particular Ct. Instead, you need to do a reference gene study to determine the stability of your reference gene(s). I see that you stated a range of 20-25 for GAPDH, that is up to 5 cycles of difference which is 32-fold (considerable). That could be due to differing RNA inputs for cDNA or cDNA dilutions, but what if there was biological variability in GAPDH expression? You would not be able to answer that if you did not do the same ng input and same cDNA dilution. You should think about whether the deltaCt or deltadelta Ct is appropriate for your experiments, but I can tell you that for Livak's method you will need to make sure you run a reference gene stability test. I will leave the original paper for you to have a read through Article Analysis of Relative Gene Expression Data using Real-Time Qu...
And never assume some reference gene is stable unless it's been empirically shown to be so in your experimental conditions, I've often found GAPDH and Beta-actin to work in one group but not in another.
- Badges
- Science method