HBDA plugin to process molecule as such??

Hi again!

As we struggle to upgrade our pharmacophore flagging issues, I was pleased to see that the HBDA plugin has some cool behaviors - it does NOT consider ester or diphenyl ether O as acceptors (however, it does consider nitro O - are you sure??). The point is that... I could not find any description on what is and what is not supposed to be an acceptor or donor according to this plugin - I had to figure these ones out by good old experimental methods...

Also, I'd like to use the HBDA plugin like exemplified in pharma_calc.xml; to do pharmacophore flagging BUT I'd like the plugin to stop checking out microspecies, and take the molecule I give as it is (this happens to be a microspecies, for I have an explicit loop flagging EACH microspecies as a standalone compound). So, I want the plugin to say that in "CCN" N is an ACCEPTOR/DONOR, not CATION/DONOR (the ammonium version of the amine will be sent for flagging at the next iteration). So, in the .xml specification I wrote

    <Plugins>
      <Plugin ID="acceptor" class="chemaxon.marvin.calculations.HBDAPlugin">
        <Param Name="type" Value="acc"/>
      </Plugin>
      <Plugin ID="donor" class="chemaxon.marvin.calculations.HBDAPlugin">
        <Param Name="type" Value="don"/>
      </Plugin>

i.e. I zapped the pH setting lines of your original example. But that still won't do it - I saw that the interactive version has a button "show microspecies data by pH" and if I check it off, I get what I want. Also, the API has some boolean to set using microspecies. But... how do you write that in PMapper .xml language?

Cheers!

Dragos

Hi,

 

it does consider nitro O - are you sure??

Yes, I'm sure. The two "O" atoms must be acceptors.  The  "nitro group" has H bonding ability according to the relevant literature. 

Somebody else will respond for the rest  of your questions.

 

Jozsi

Mmh... I agree that lately I posted a lot of stuff and got very quick answers, but actually this is the most urgent issue to us: turning off the microspecies enumeration within the HBDA plugin in pmapper. Can you please tell me the magic xml spellword to add to the pmapper config file, so we may get on with our descriptor calculations? Köszönöm szepem!!

Hi Dragos,

I guess this will work:

<Plugins>
  <Plugin ID="acceptor" class="chemaxon.marvin.calculations.HBDAPlugin">
    <Param Name="type" Value="acc"/>
    <Param Name="mscalc" Value="false"/>
  </Plugin>
  <Plugin ID="donor" class="chemaxon.marvin.calculations.HBDAPlugin">
    <Param Name="type" Value="don"/>
    <Param Name="mscalc" Value="false"/>
  </Plugin>
</Plugins>

pmapper developers will correct me if I'm wrong.

Sorry for the delay.

Zsolt

Well, this is in fact exactly what I had tried out spontaneously, saying to myself "if I were a chemaxon developer, I'd call that variable 'mscalc' and set it to false". Legalize telepathy!

Small hiccup - it does NOT work that way! Joined are the typing program and the xml file; if you "echo CCN|java Utils/Type" the pharmacophore flag of the N in the resulting output Typed.sdf should be "A/D" and not "D/P" (P - positive charge).

Oops - sorry, actually it DOES work! I had just mismatched the code snippets - the one in which I had outcommented the microspecies enumeration mechanism in order to stick to the molecule as input was called "Type1", not Type.java - and it does what it should do! Everything's fine - thank you!!