Virtual Screening

Dear Sir


I was trying out the examples mentioned in the index.html file under the topic of virtual screening


I got errorous results in computation against the example mentioned for the optmization of hypothesis when executing the command. I have attached the files for your kind reference.


 


hitstatistics nci700.smiles ace-9-16.smiles ace-5-8.smiles -o hitstat.txt -e 3 -H Median -b -k PF -c pharma-frag-opti.xml


FILE. hitstat.txt


 


and


 


optimize nci1000.smiles ace.smiles


FILE. example_statics.stat

Hi,





known issue that might be related to the problem you encountered. You may try the -f option in the command line of hitstatistics, that can solve your problem. Please get back if that does not help.





Regards,





Miklos

Thank you Sir





the -f option has helped to generate the hitstat file, but still the results do not match the example results in the readme of the virtual screening.  I tried the -f option in optimize command its not functional





Please help me out











Thank you





Raghuvir

The -f works with hitstatistics only, the optimizer does not recognise this option.


Can you attach the output file you got? That would help us better understand the problem.





Thanks,


Miklos

Dear Sir





I have attached the output file of optimize





Thank you





Regards





Raghuvir

Isn't the hitstat output file attached to your first post the one that was not produced properly? I meant the one that was generated by the use of the -f flag.














Thanks





Miklos

Dear Sir





I am attaching two files for your references, the outputs of hitstats without -f flag and with -f flag.





however the results do not match the results depicted in the example.





Also how one can use the optimize.bat file





regards





raghuvir

Dear Raghuvir,





I confirm that the output you received is produced by version 5.0 onward. I have not yet managed to trace back what causes the difference in the output. 





I do not suspect any major error in any tools involved in this procedure, yet, it would be reassuring to figure out what causes this difference. I will get back to this forum with my findings. In the meantime, however, you can use the tools and trust in the results: values may change yet the rank of the metrics should be consistent.





Apologies for any inconvenience this might cause.





Regards,





Miklos