Regarding use of fragmenter recap algorithm

Dear all,

I have seen that fragmenter for windows gives us two options

 

a) using fragmentAlll.xml and

b) FragmenterRecap.xml

but in both the experiments input file is different.

I have done fragmentall.xml using .sdf file format for 50 structures along with its activity value. But since my structures are aromatic and heterocyclic nucleus hence i will have to use recap also for all the compounds.

now FragmentRecap only uses .mol file which for one structure at a time. And it is impossible for me to use if for large number of structures.

please suggest me how to convert activity data for this large structure dataset for fragmentation analysis using fragmentrecap?  or u think only fragment statistics with fragmentall can solve the problem? As i dont want to do harsh fragmentation  because it may cleave the rings also...

Waiting for your reply

 

Thanks,

Ankit

Dear Ankit,

Sorry for getting back to you so late.

FragmenterAll.xml and FragmenterRecap.xml are two example configuration files that you can use with Fragmenter to get started, but you are not restricted to use these configurations. You can modify them or even writhe your own configuration files. If you open these files with a text editor, you will find that the fragmentation actions that are applied by them are pretty much the same. These are defined in the Actions section. FragmenterRecap.xml will apply a few additional, so called revision rules on top of the original RECAP rules. You can find these in the Reviser section of the file. To find out more about these rules please visit http://www.chemaxon.com/jchem/doc/user/fragment_recap.html

As for the rings: the fragment method (RECAP) *never* cuts ring bonds, so you don't need to be concerned about the algorithm destroying the ring cores of your molecules.

Regarding the file formats: sdf files can be used as input regardless of which configuration file you use. This means that even if you take FragmenterRecap.xml, you are not restricted to .mol files. You can apply this configuration file on your large structure collection just as well as FragmenterAll.xml. The example on our webpage may be misleading in this respect, but there is no such restriction at all.

If you want the activity data from your input files to be in the focus of your fragment analysis, you will need to apply (as you wrote correctly) FragmentStatistics after the fragmentation step is done. Just as described above, you can do that with FragmenterRecap.xml. In order to do so, you need to make sure that the output format is CXSMILES. I'd like to point your attention on the corresponding documentaiton: http://www.chemaxon.com/jchem/doc/user/FragmentStatistics.html

 

I hope this helps,

best regards,

Ivan