You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks for the great cell-cell correlation function. Do you mind tell me which method we used for the correlation analysis, eg Pearson, Spearman, etc? Also, we don't quite understand why
if (PVal == 0) {
PVal = "2.2e-16"
}
Is it the reason our PVal is very small?
thanks,
Yale
The text was updated successfully, but these errors were encountered:
Thank you, yes! It's the general r "cor.test" function that by default takes "pearson". Thanks for reminding me, I will add the option to change methods in the future. It is a bit of a lazy coding but yes.
I think at the time I wanted to replace 0 with the smallest p value because for some reason R was converting them to 0. But I will look into it in the future. Generally they are also represented with 4 stars (****) for p < 0.00001. If you want the exact very small p number you can run the "cor.test" outside of the function on the imputed data.
You are right, they always report 0 if it is very small. Your code seems to keep the real p-value, right? As I always get XXe-267 or XXe-578, etc.
Also, I compared Pearson and Spearman, it seems Spearman is more suitable for single-cell data, as it considers the ordinal variables. Then can you tell me why you set the default Pearson?
Hi iCellR,
Thanks for the great cell-cell correlation function. Do you mind tell me which method we used for the correlation analysis, eg Pearson, Spearman, etc? Also, we don't quite understand why
Is it the reason our PVal is very small?
thanks,
Yale
The text was updated successfully, but these errors were encountered: