p-value on excel for different number of mutant samples vs WT samples

Question

[abhatnagar]

I have an RNA-seq data set for with 6 mutant gene sample values and 5 wildtype control values. How do I conduct a two-tailed test to gain a p-value for this data set on excel? I have already normalized the values and found both the fold change and the log2 fold change so far. I'm having issues calculating the p-value since the number of samples from mutant gene is not the same as the number of values from wildtype. How can I do this? Any advice is appreciated, thanks for taking the time!

Answer 1

[ATpoint]

Naive tests are not appropriate for RNA-seq with low sample size, neither is Excel. Use specialized packages such as DESeq2 in R.

[abhatnagar]

Thank you for the reply! I'm a very low-level beginner at R, which is why I was using excel, but how do I use DESeq2 in R?

[ATpoint]

Read its manual. It covers everything. Or get local help from someone with R knowledge.

[abhatnagar]

Thanks! I think I may have conveyed my dataset differently: I have a large sample size of about 21k values in each set but I have 6 sets of samples for the mutant and 5 sets of sample for the wildtype. I would like to do a preliminary analysis via excel so I have something for my meeting, would you know how?

[ATpoint]

I have no other advise than what I already said. 6 vs 5 is low sample size. Naive analysis is unreliable so you're probably discussing artifacts in your meeting.

Nobody here will provide you Excel help for such things so get this out of your head and spend the time learning DESeq2 usage. It's few lines of code and runs quickly.