On Sat, Mar 19, 2022, David Poole wrote:
>I found success in getting it installed using conda forge independently of
>my amber/ambertools installation. I am unsure why it didn't work off the
>bat as parmed.py and pytraj both worked (mostly), and even pamred.py
>allowed me to use esander. I think it was just a matter of the version
>included in ambertools?
I'm confused by the statement "getting it[?] installed using conda forge".
Do you just mean "conda install -c conda-forge ambertools=21 compilers"?
You also say "it[?] didn't work off the bat": what changed from before?
We're happy to take suggestions about the documentation on github.
Are you referring to the information at
https://github.com/Amber-MD/pytraj?
Mainly, we expected that users would get all needed documentation from the
Reference Manual, and just use "conda install" as an alternative to building
the codes from source.
>
>Alas now that things are working, I have a bug to report: While I can use
>esander in pytraj to do gb (igb > 0) and pme calculations, when I attempt
>to do two or pb calculations (ipb > 0 or igb = 10), I get a
>segmentation fault from pb_init() and the program exits. The minimum
>example is computing the single point pb for any two loaded structures in
>pytraj. Calculations can be set up with either mm_options (options.ipb > 0,
>or igb = 10) or by setting igb in the esander method (igb = 10), which all
>lead to the same result. This would be particularly useful for my new
>workflows for mmxbsa calculations using the pytraj interface.
Hai is the best person here, and the above sounds like this issue:
https://github.com/Amber-MD/pytraj/issues/1605
If so, it's probably best to keep the discussion there. I see that Hai has
some detailed suggestions from just 2 days ago, so maybe those will help.
...thx...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Apr 05 2022 - 07:00:03 PDT
