Diverse subset selection

User 8a73bbc4d4

15-07-2014 11:21:07


Hi,

I am trying to cluster a 30000 database based on the ECFP dissimilarity. I want to get 10% of the most representative compounds, using K-means cluster method. Although, when I type:

/Applications/ChemAxon/JChem/bin/jklustor USEF_chemical_library_std_smiles.smiles -c kmeans:3000 -d ecfp > results.txt

It stays there forever and I never get an output. What am I doing wrong?



Ana

ChemAxon 8b644e6bf4

18-07-2014 16:54:01

Dear Ana,


 


Would the sphere exclusion method suitable for the diverse selection?


 


Regards,


Gabor