Fragmenting the CC bond

Hi

I am interested in fragmenting the CC bond. As per the RECAP rules we cannot get a CC cleavage but with the CCQ we can. I am interested in the pdb output but with the CCQ I get zero coordinates. Can you suggest me what changes I can make in the RECAP xml (i.e.  FragmenterAll.xml ) so that I can get a CC cleavage. I was trying with <Action ID="carbon-carbon" Structure="[C:1]-[C:1]>>[C:1].[C:1]"/> which gave me the cleavage at CC but the problem is I can't get the fragement of interest even after specifying x and y as 1000.

 

regards

pramod

Dear Pramod!

Could you please specify more detailed what kind of additional datas you need for the output? I'd like to ask you to give us some example of the input you have and the output you would like to get from the fragmanter tool. (pdb/mrv or some same format with coordinate data would be very helpful)

Attila

Hi Attila,

       I expect all possible combinations of fragements with a cleavage between C-C bond. For example I have a molecule POPC (attached) I expect C-C cleavege and all possible fragment combinations. I get the C-C cleavage with the coordinates (for output fragments) when I use

<Action ID="carbon-carbon" Structure="[C:1]-[C:1]>>[C:1].[C:1]"/> but the problem is the fragment combination are not enough even when I specify x any y as 1000 (I get around 20 fragments)

 

For example when I use this command

fragment -c Fragmenter.xml popc.pdb -x 1000 -y 1000 -f pdb -o out.pdb

I get only 20 fragments

where Fragmenter.xml specifies

<Action ID="carbon-carbon" Structure="[C:1]-[C:1]>>[C:1].[C:1]"/>

So  how can I get all the fragments? and

how can I filter and remove the unwanted fragments (like smaller ones) ?

regards

pramod

The main problem here is that there are 36 CC bonds and hence there are 2^36 possible cut-sets which takes far more time to process than one can wait for. However, to generate all possible fragments (instead of all cut-sets) we can make some reasonable restrictions on the cut-sets, based on the structure of the input molecule.

 

Observe that the molecule has nearly a 3-branched tree structure as long as CC bonds are considered. Therefore we can set a maximum number of 3 cuts per fragment set (-x 3) and still generate all possible fragments.

However, use extensive mode (-e) to include those sets that can be extended further by more cuts.

With these settings, we can expect nearly 8000 fragments but with repetitions. I did not count the different fragments but there should be far less than this.

I converted the input to 2D and generate output in smiles:

fragment -c Fragmenter.xml popc-download5942.mrv -f smiles -e -x 3 -o o.smiles

This produces 410 fragments.

 

Now to filter small fragments I can think of 2 ways, both uses RECAP parameters:

1. set MinAtomCount

2. set NotList to exclude trivial fragments

 

Example:

fragment -c FragmenterRecap.xml popc-download5942.mrv -f smiles -e -x 3 -o o.smiles

This produces 262 fragments.

 

 

 

 

 

Dear Nora,

       Thanks for the reply and your effort. It worked out well.

Where can I get the detailed information on writing configuration file and modifying as per my convenience? Also I want to know the complete information on fragmenter.

Cheers!

pramod

Dear Pramod,

here are the links to documentation:

• Fragmenter main doc


• Fragmenter with / without RECAP in more detail


• Configuration


• Examples

 

Best Regards,

Nora