JChem and Proteins?

Hi!





Is it possible or rather advisable to use JChem with bik Molekules like Proteins? I've seen the ligand search system at rcsbPDB amd want to implement an similar Service.


And if its possible, i know that i've to convert the PDB's with the molconverter.. but which format is the best before importing them with Jchembase?


If it's not advisable, or possible i would be intrested how the rcsbPDB solved this prob...





Thanks for help..





Björn

Hi Björn,





Unfortunately, JChem Base/Cartridge is not the best system to store large proteins. Not only because of the size, but because of the high similarity of protein building blocks, it is very probable that fingerprint screening is not very efficient. (Smaller polypeptides should be OK though.)





The PDB guys separated the ligands from the PDB and converted them to sdf format, I believe.





I hope this helps,


Szabolcs

For the separation of ligand and protein fragments, our Standardizer tool can help. See the <Removal> action. (On the Standardizer GUI, it is called "Remove fragment".)





This action with the "removeLargest" option could be used repeatedly to remove the protein chains from your PDB file.





I hope this helps,


Szabolcs

Thank you for your fast reply and your suggestion. I've tryed the standardizer with some randomly choosen PDB file and became reproducible an Arrayindex out of Bound error or an Invalid Bound(0,0). I used a very simple conifig:





<?xml version="1.0" encoding="UTF-8"?>


<!-- Standardizer configuration file -->


<!-- This configuration file is created with ChemAxon Config Builder -->





<StandardizerConfiguration Version ="0.1">


<Actions>


<Removal ID="removal" Method="removeLargest"/>


</Actions>


</StandardizerConfiguration>





What do you think of the way to extract only the HETATOM data from the pdb-file and using JChhem with them?





with best regards


Björn

Hi Björn,





- HETATM only should be o.k.





- i'll try to reproduce the problem you experienced in standardizer and i will get back to you and let you know when the bugfix release is available





- are you happy to do some Java programming using our API to solve your problem, or you prefer to use readily available tools? In this latter case we can, for instance, implement a new standardizer action: keep ligand, or remove protein etc.





Thanks and regards,


Miklos