On Fri, Apr 01, 2022, Erdem Yeler wrote:
>I parameterized Pd atoms with almost all known bonds. For this, I wrote a
>script to combine gaff and a novel metal force-field (the script can be
>adapted with Pt, Au, Se, exc...) Everything works perfectly, but there is a
>mini problem: Sander minimized the structure with very high accuracy, but Pd
>lost its bonds and seemed like a "P" atom when I visualized it (I changed
>the "Pd" sign to "Pm" (because in gaff there is pd even if it it is not
>exactly Pd with capital P) but I couldn't solve the problem.) Why is this
>happening? Actually, I don't see it as a serious problem because the
>minimized geometry is correct, but if there is a way to get rid of this, I
>am ready to apply.
>Here is the picture of my molecule: P must be Pd and must have bonds with
>Cl and N atoms. I checked the prmtop file with parmed. PArmed seed Pd atom
>and related bonds but after ambpdb commands it seems bond-less P atom...
This looks like a visualization issue; ambpdb does not output CONECT records
unless you use the '-conect' flag. So, the whatever program you are using
has to guess about what bonds to draw based on interatomic distances.
If parmed says the bonds are there, then the bonds are there. And you would
not get a good geometry if they were absent. Try adding the -conect flag,
or save your output to mol2 format (which as bond information), assuming
that your visualization program supports the mol2 format.
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Apr 01 2022 - 05:30:02 PDT
