p value in box whisker

Question

[rob.costa1234]

A very naive question I want to show significance differences in distribution of two tests of values, I an easily show by box plot and have confidence of interval calculated. I have see several papers also showing a p value. How can I calculate p value to show differences among two box plots. A google search showed that CI an p values are calculated from same formula. Then how can I get p value either from CI or from a different calculation from two distributions

Thanks

[jotan]

The box plot is only a way of displaying data.

Statistical tests are based on the underlying dataset. What kind of data do you have?

[rob.costa1234]

I have gee expression between disease vs control

[jotan]

Across multiple genes? The more information you give, the more people can help.

Is it something like this?

Fpkm Comparison Using T-Test Or Mann-Whitney U Test

[rob.costa1234]

4 groups with 10 replicates : two normal groups untreated/ treated; two groups of disease untreated/ treated . DE was calculated using ANOVA.

Answer 1

[Ron]

You can calculate a p-value between the two box plots by doing a t-test of all genes of one group vs other. You can combine the two normal groups into one and two tumor groups into one,and then do the t-test .