Re: about paired or unpaired t.test or wilcox.test in R
On 9/11/19 1:14 PM, Bogdan Tanasa wrote:
> Dear all,
> if would be great if you could please advise on the use of PAIRED or
> UNPAIRED T.TEST or WILCOX.TEST in R :
> let's say shall we have 2 samples :
> -- CONTROL : where we measure the expression of 100 genes G1 ... G100 in
> one million cells C1 ...C1mil
> -- TREATMENT : where we measure the expression of 100 genes G1 ... G100 in
> one million cells D1 ...D1mil
> when we compare the expression of these 100 genes G1 ...G100, in CONTROL vs
> TREATMENT, shall we use UNPAIRED TESTS, correct ?
> as the cells in CONTROL C1..C1mil are different than the cells in TREATMENT
> D1..D1mil ? thanks a lot !
> -- bogdan
> [[alternative HTML version deleted]]
Rhelp is a plain text list. It is specifically for help with R coding
difficulties, and it is specifically NOT to advice about chosing
analysis plans. You question raises serious analysis issues regarding
multiple comparisons and distributional assumption which I believe are
still the subject of research. A more welcoming venue might be the
BioConductor mailing list or CrossValitdated.com.