Hello,
I am looking at protein expression across mouse retina. We use littermate controls so many times our sample size is limited to 4-8 samples. To ensure protein does not degrade, we usually lyse the samples as soon as possible and run them immediately on gels. Because we are limited by how quickly the mice can breed, we are limited to one litter at a time, so our protocol has been one litter one gel. After this we run the experimental probe, and then load control of Beta-Actin. We then do ImageJ quantification and densitometry and take experimental probe/ load control and get normalized expression.
My questions are:
What is the correct way to compare the normalize expression between gels?
You can model the blot (gel, litter) as blocking variable. If you have many blots (more than 5-10, say), employing a mixed model with "blot" as random intercept could be a better approach.
If you have enough sample, you could run the same sample on each blot to serve as a control. This control could be from a single animal/litter, or a pooled sample comprising a small amount from each animal in the study. The drawback of using the latter is that you have to have to wait until you have all of your samples on hand.
- Badges
- Science method
- Similar topics
- Biological Science
- Genetics
- Gene Expression
- Protein Expression