openmolecules.org

 
Home » DataWarrior » Bug Reports » Pentavalent carbonyl issue (urea formation?) (Enumerating combinatorial library)
Pentavalent carbonyl issue (urea formation?) [message #1938] Thu, 22 June 2023 21:14 Go to next message
padster is currently offline  padster
Messages: 5
Registered: June 2023
Junior Member
Hi, I am trying to create a combinatorial library (macrocyclic peptide) in DataWarrior. The issue I run into however is that every amide bond forms correctly with the next amino acid, but for some reason it also forms a bond (red bond in image) between the N and C=O of the same amino acid, creating a urea-like motif (refer to image attached). If the red bond is deleted, it forms the correct structure. Curiously, this issue does not happen when running DW on Mac. Is there a way to fix this on Windows so that it runs correctly as well, or is there some sort of setting that changes this? I reinstalled DW, but still run into the same issue.

Macrocycle: /forum/index.php?t=getfile&id=727&private=0
Reaction: /forum/index.php?t=getfile&id=728&private=0
Re: Pentavalent carbonyl issue (urea formation?) [message #1939 is a reply to message #1938] Fri, 23 June 2023 09:26 Go to previous messageGo to next message
thomas is currently offline  thomas
Messages: 666
Registered: June 2014
Senior Member
Can you post the reaction as .rxn file, please? Or better a small macro containing the complete combinatorial library task with reaction and one real reactant for each generic reactant? This way I could directly reproduce the issue. Thanks in advance...
Re: Pentavalent carbonyl issue (urea formation?) [message #1940 is a reply to message #1939] Fri, 23 June 2023 20:02 Go to previous messageGo to next message
padster is currently offline  padster
Messages: 5
Registered: June 2023
Junior Member
Sure thing, attached are four .dwar files with one structure each for one cycle and the .rxn file:

I'm not sure how to provide a macro to be honest

[Updated on: Fri, 23 June 2023 20:22]

Report message to a moderator

Re: Pentavalent carbonyl issue (urea formation?) [message #1941 is a reply to message #1939] Fri, 23 June 2023 20:27 Go to previous messageGo to next message
padster is currently offline  padster
Messages: 5
Registered: June 2023
Junior Member
Perhaps this macro works?
Re: Pentavalent carbonyl issue (urea formation?) [message #1942 is a reply to message #1941] Sat, 24 June 2023 16:53 Go to previous messageGo to next message
thomas is currently offline  thomas
Messages: 666
Registered: June 2014
Senior Member
Thank you for the files. To me it seems that your atom mapping is incorrect. The reactants fragments consisting out of a carbonyl group with attached bridge bond to nitrogen should not break bonding in the reaction, thus they should appear in the product the same way with the same mapping numbers. I have attached the updated macro file, which seems to work fine now.

The upper reaction shows your original mapping, the lower one shows the updated mapping:

/forum/index.php?t=getfile&id=739&private=0

[Updated on: Sat, 24 June 2023 16:54]

Report message to a moderator

Re: Pentavalent carbonyl issue (urea formation?) [message #1943 is a reply to message #1942] Sat, 24 June 2023 23:45 Go to previous messageGo to next message
padster is currently offline  padster
Messages: 5
Registered: June 2023
Junior Member
Hi Thomas, thanks for getting back to me. However, I still get the incorrect product when I run your macro. When I try to make it myself, it rearranges the mapping so that each nitrogen is 1, 6, 9 or 12 and the first carbonyl is 2-3, but the second one is 4-5 (see attached image). It is also strange how my old mapping does not work on Windows, but it worked perfectly fine on Mac. I'm getting the idea that the Windows version is reading the mapping different or automatically changes it somehow (if that's even possible).

Image: /forum/index.php?t=getfile&id=740&private=0
Structure: /forum/index.php?t=getfile&id=742&private=0

[Updated on: Sat, 24 June 2023 23:47]

Report message to a moderator

Re: Pentavalent carbonyl issue (urea formation?) [message #1944 is a reply to message #1943] Sun, 25 June 2023 13:31 Go to previous messageGo to next message
thomas is currently offline  thomas
Messages: 666
Registered: June 2014
Senior Member
If I import and run the macro from my last message, then I get the following product (with both versions 5.5.0 and with the current dev update):

/forum/index.php?t=getfile&id=743&private=0

If you use the same version on MacOS, Linux, or Windows, then it should behave exactly the same. Which version do you use anyway? The official 5.5.0, that mean the most recent (but 2 years old) official installer? Then I would recommend to update to the newest dev build (on the download page click on the 'read and understood checkbox, then click the link in the small print for you platform and download the respective archive with patch files for your installation)

Did you manually map as many atoms as it needs for auto-mapper to correctly map all remaining atoms of the reaction? If not, I suggest to remove all manual mapping and start from scratch to map at least all four O=C...N fragments manually.

Another tip: Typically, one does not use atom or bond query features on the product side of the generic reaction. Bridge bonds are an exception, because whatever matches on the reactant side needs to be reconstructed on the product side. If all your real reactants have the same bridge length, then I would remove the bridge bond from the reaction and replace it with real bonds and atoms.

One more tip: A part of the new medium sized ring comes from unknown components. For the sake of complete and perfect mapping and a stoichiometrically complete reaction you might introduce a fifth reactant with the missing atoms, that is always the same.
Re: Pentavalent carbonyl issue (urea formation?) [message #1945 is a reply to message #1944] Sun, 25 June 2023 20:56 Go to previous message
padster is currently offline  padster
Messages: 5
Registered: June 2023
Junior Member
Installing the dev build seems to have done the trick! Thanks a ton!
Previous Topic: conformer generation sometimes skips isomer generation
Next Topic: conformer generation misses some entries
Goto Forum:
  


Current Time: Wed May 22 06:25:45 CEST 2024

Total time taken to generate the page: 0.04157 seconds