**330**wrote:

Dear group,

I use RnaSeqSampleSize package for calculating power analysis.

Documentation for this package suggests the following usage:

sample_size(power = 0.8, m = 20000, m1 = 200, f = 0.1, k = 1, w = 1,rho = 2, lambda0 = 5, phi0 = 1, showMessage = FALSE,storeProcess = FALSE)

where k = Ratio of sample size between two groups.

In the case of an unbalanced design (e.g., 70 in group A, and 30 in group B), how k should be defined. (k = 70/30 = 2.333)

sample_size(power = 0.8, m =15000, m1 = 400, f = 0.01, k = 1, w =1,rho = 2, lambda0 = 10, phi0 = 0.5)

[1] 50

sample_size(power = 0.8, m =15000, m1 = 400, f = 0.01, k = 2.3, w =1,rho = 2, lambda0 = 10, phi0 = 0.5)

[1] 3

As shown above sample size of 50 and 3 for k = 1 and 2.3 respectively is not correct.

Perhaps, I am not defining k properly. Could anyone help understand this. I appreciate your help.

Thanks!