Hello and thank you for making this very useful package.
I was wondering if it is possible to use the est_power_curve function with a user's prior dataset?
I see that this function does not take distributionObject as input. I was wondering if it would be appropriate to take the dispersion estimate calculated for my data using the est_count_dispersion function and input it as the phi0 parameter for the est_power_curve.
I am not sure if that is a sound approach since the phi0 parameter of the est_power_curve function is defined as "Dispersion for prognostic genes." and the est_count_dispersion dispersion output seems represent the dispersion for all genes included in the analysis.
Alternatively I thought about using a loop to calculate power for different sample sizes with the est_power_distribution function using my prior dataset (this function takes a distributionObject parameter). However, that is a bit slow and computationally intensive for the use case that I have in mind.
Any help or guidance here will be much appreciated.
Hello and thank you for making this very useful package.
I was wondering if it is possible to use the
est_power_curvefunction with a user's prior dataset?I see that this function does not take
distributionObjectas input. I was wondering if it would be appropriate to take the dispersion estimate calculated for my data using theest_count_dispersionfunction and input it as the phi0 parameter for theest_power_curve.I am not sure if that is a sound approach since the
phi0parameter of theest_power_curvefunction is defined as "Dispersion for prognostic genes." and theest_count_dispersiondispersion output seems represent the dispersion for all genes included in the analysis.Alternatively I thought about using a loop to calculate power for different sample sizes with the
est_power_distributionfunction using my prior dataset (this function takes adistributionObjectparameter). However, that is a bit slow and computationally intensive for the use case that I have in mind.Any help or guidance here will be much appreciated.