Cxcalc LEConformer - ginormous output file size?

Hello,

I ran from the shell the ff. command:

./cxcalc -S -v *.smi -o filename leconformer

I was wondering why the current output file size for around 4.5 million compounds (in SMILES format, and the only *smi file in the directory, 250 MB size uncompressed) is already around 50 GB!

Does it store all conformers? Or how many conformers on average is written to the file? I was thinking it was only one LE conformer, but looking at the file size, I am stumped.

I have not been able to find any flags for cxcalc that can improve this.

Thanks,
Enrique

Hi, 

Thank you for reporting this. Yes, the leconformer function should give back just one conformer with the lowest energy.

However, we would need a test set so that we can re-produce the issue.

You can view all available options of the lowest energy calculation with the

cxcalc leconformer -h 

command. 

I hope this helps, 

Daniel

Thank you for your prompt response. I actually got it from: http://zinc.docking.org/db/bysubset/11/11_p0.smi.gz This is the ZINC clean leads subset. Cheers, Enrique danielszisz wrote:

Hi,  Thank you for reporting this. Yes, the leconformer function should give back just one conformer with the lowest energy. However, we would need a test set so that we can re-produce the issue. You can view all available options of the lowest energy calculation with the cxcalc leconformer -h command.  I hope this helps,  Daniel

Hi, 

Thank you for sending the input file. We will investigate this issue as fast as our priorities allow us. 

We will keep you updated via this forum. 

Best regards, 

Daniel