purity=1 bug - Githubissues

Wedge-lab / dpclust

Dirichlet Process based methods for subclonal reconstruction of tumours

GNU Affero General Public License v3.0

28 stars 16 forks source link

purity=1 bug #13

Open kamichiotti opened 2 years ago

kamichiotti commented 2 years ago

There is a bug causing a function to divide by 0 when purity is equal to 1. It can be worked around by changing purity=1 to purity=0.999.

The error generated by this bug is:

Error in density.default(c(pi.h.cols[post.burn.in.start - 1, ]), weights = c(wts[post.burn.in.start,  :  non-finite 'to' 
Calls: RunDP ... DirichletProcessClustering -> <Anonymous> -> density -> density.default
Execution halted

Avramis commented 2 years ago

Hi @kamichiotti . Sorry for the late reply. Would it be possible to provide more details so we can try and replicate the issue?