# Thread: Statistical Test for fold changes of gene expression

1. ## Statistical Test for fold changes of gene expression

Hi everyone,

I am currently doing my masters and working on gene expression in tumor and control animals. I am really lost when it comes to the right statistical tests and hope you can help me.
I meassure the expression in a specific cell type and (for simplicity) just have two groups: "control" and "tumor". I repeated the experiment 4 times - always with littermates from the same mother, half of which got a tumor and half did not. I use quantitative RT PCR and a delta delta Ct calculation which gives me fold changes in the end.
So in the end i get something like this:
Control: 1 , 1 , 1 , 1
Tumor: 14 , 50 , 30 , 100

My questions are the following:
- Do i have a paired or unpaired observation here?
Each mouse is a different animal but i matched them so I can compare between the two groups and I calculate the fold change for each experiment so they have to be seen together (1 and 14, 1 and 50, etc)?

- Which statistical test could I use to show that I always have a strong induction of gene expression in the tumor group?
The high variation gives me no statistical significance with a paired t-test but I'm not sure I'm allowed to use that one.
A colleague suggested a Mann-Whitney-U Test but I'm not sure either whether I am allowed to use that one here.

I would be really glad I you could give me some suggestions!

biba

2. ## Re: Statistical Test for fold changes of gene expression

I suggest Mann-Whitney-U Test for your data.

3. ## Re: Statistical Test for fold changes of gene expression

As far as I understand Mann-Whitney-U is for unpaired data - is that the case for my "fold change" data? Since one column is always expressed relatively to the other one.

4. ## Re: Statistical Test for fold changes of gene expression

hi,
I am not sure that the prerequisites apply, but maybe you could try a two-way ANOVA? One factor could be treated or not treated, the other the litter. So, you could isolate the effect of the litter.

The fact that anpaired t did not give you a significant result is not a good sign though.

regards
rogojel

5. ## Re: Statistical Test for fold changes of gene expression

This is next generation sequencing data right? Are you only interested in a single gene or are you going to be running this test for multiple genes?

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts