Hi Viktor,
Thanks for the effort. We are facing a new release of AmberTools in the
next month or so. There are a lot of changes, and at least pthread for
memembed is already in there ;D. For pdb4amber, I think you might be
partially right. The issue is that in many pdbs, the element entry is not
there, or is used for something else. Probably something like (@H= | @/H)
would cover the whole spectrum. CC to Hai to see what he thinks.
Cheers,
El mié., 10 abr. 2019 a las 13:28, viktor drobot (<linux776.gmail.com>)
escribió:
> Hello, dear community! I've prepared two patches - for memembed Makefile
> and for pdb4amber.py utility.
>
> The first patch fixes problems with linking process of memembed
> (actually adds -lpthread to LIBS).
>
> The second patch fixes incorrect behavior of --nohyd option of pdb4amber
> utility. Following /TRPCage tutorial/ we've found that PDB file with NMR
> structure which is supplied in the /section 6/ (*nmr_struct_1.pdb*,
> *http://ambermd.org/tutorials/basic/tutorial3/files/nmr_struc_1.pdb*)
> contains hydrogen atoms which names start with digits (perhaps VMD
> yields incorrect PDB). Digging inside pdb4amber code has shown that mask
> for hydrogen stripping is based on atom name rather than element type.
> So I've suggested that using element type-based mask is better. I've
> tested it on the PDB file mentioned above and things look great. Patch
> is attached here.
>
> Thank you all for your work!
>
>
--
Stephan Schott Verdugo
Biochemist
Heinrich-Heine-Universitaet Duesseldorf
Institut fuer Pharm. und Med. Chemie
Universitaetsstr. 1
40225 Duesseldorf
Germany
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Apr 10 2019 - 06:00:05 PDT
