Normalization of gene expression data

Question

[bioinfonerd]

Hi everyone,

I have 50 samples gene expression data and I normalized them together. Now I want to use just 20 samples out of these for my analysis, do I need to normalize the raw counts of these 20 samples again or I can use the normalized values from the normalization (50 samples norm) that I did before?

Thanks

[bioinfonerd]

Thanks a lot! Could you also let me know the advantages of doing that?

[WouterDeCoster]

Please use ADD COMMENT or ADD REPLY to answer to previous reactions, as such this thread remains logically structured and easy to follow. I have now moved your reaction but as you can see it's not optimal. Adding an answer should only be used for providing a solution to the question asked.

Say that you have a few outliers in your samples. The result of these on the normalization will be bigger when you only have a few samples. When you have more the effect will be diluted.

[bioinfonerd]

Thanks that was really helpful!

[Kevin Blighe]

If an answer was helpful you should upvote it, if the answer resolved your question you should mark it as accepted.

Answer 1

[WouterDeCoster]

If all samples were processed on the same platform/same kit/same lab the best normalization you can obtain is to include as many of them as possible, so in this case normalize using 50 samples and only use 20.