batch_sil with non-numeric batch codes

[jesswhitts]

Hello,

I get the following error when running batch_sil:

R[write to console]: Error in if (!all(x == round(x))) stop("'x' must only have integer codes") :
missing value where TRUE/FALSE needed

R[write to console]: In addition:
R[write to console]: Warning message:

R[write to console]: In silhouette(as.numeric(batch), dd) :
R[write to console]: NAs introduced by coercion

The problem comes from having non-numeric batch codes, which this line of code converts to NAs:
summary(silhouette(as.numeric(batch), dd))

Can you suggest any workaround for this?

Many thanks,
Jess

[mbuttner]

Hi @jesswhitts
thanks for pointing this out! A quick fix would be converting all batch codes to numeric values. If your batch covariate is a factor, you can usually convert the values of a factor to numeric with as.numeric(batch), as shown in the code. You can check if using a factor fixed the problem. I cannot reproduce the error, so it would be helpful to have a little example along with your R version.
Thanks!