Hi Prof. Case :
Thanks very much for replying and for the suggestions-- I've tried each of these (first, a "fresh" installation of AmberTool17 into its own tree (headed by ../amber16) and pointing $AMBERHOME there). I'm afraid this leads to the same problem.
Merging the AmberTools17 tree into my amber14/ installation (both with and without removing the existing AmberTools) also did not help. A fresh recompile of amber14 with Amber17 present did not help either.
Finally (tried today) a wipe of All Things Amber from a test machine, followed by anaconda-package-based-installation of AmberTools17 as per the instructions on the ambermd.org website did not help. It *worked*, in that minab, mdnab, etc. all work fine, getting me to the same place I was before I posted this, but upon running pymdpbsa , I get the exact same failure and error message as previously reported.
Since posting my original question I've also heard from Dr. Wolf who, from his look at the error, thinks a problem has occurred over time. A new version of 'amberlite' is supposedly in the works.
While on the topic, I was wondering if there existed a script repository or the like, containing at least a starting workflow analogous to that in the amberlite & associated p38 kinase example, but using pmemd (advantageous for many reasons of course) / MMPBSA.py to perform the ligand parameterization, MD run, and ensemble-averaged MM-GBSA energy extraction?
I'm not adverse to writing up something like this from scratch as I dig deeply into the many options and flags available, but if I need not reinvent that wheel that would be helpful, at least as a starting point. In my separate communication Dr. Wolf, he made mention of an eventual release of something like this in a future version of AmberTools, but I'd be appreciative if I could find a reasonable starting point or tutorial for doing the equivalent of what is in, say, the amberlite p38 example using pmemd and other modules in our licensed amber14 as the engine.
Thanks once again for your reply,
Cheers,
-Michael Bartberger
Amgen
-----Original Message-----
From: David A Case [mailto:david.case.rutgers.edu]
Sent: Sunday, September 03, 2017 6:27 PM
To: AMBER Mailing List
Subject: Re: [AMBER] amberlite p38 tutorial: Failure of final step of "pymdpbsa" in amberlite
On Sun, Sep 03, 2017, Bartberger, Michael wrote:
>
> My group is a licensee of Amber14 (the most current version we've
> licensed) into which AmberTools 12 has been integrated (and updated
> via the "update_amber --upgrade" mechanism, last checked just
> yesterday. The update appears to have been successful.
You should certainly install AmberTools17 into its own directory tree (headed by ...amber16), and point your AMBERHOME variable there. (The only program not in AmberTools is pmemd, and you can continue to use the pmemd from Amber14: it doesn't need or use AMBERHOME).
Note also that amberlite is entirely within AmberTools, so you may get lucky if you replace AmberTools12 with AmberTools17. I'm not sure if the --upgrade option works all the way back to AmberTools12; certainly you still seem to be referring to the 2012 version of the Reference Manual, and things may have been fixed in the five years since then.
If you got to the last step of the example, see if you can figure out from the documentation what the next steps should be. (I appreciate that this is not an answer to your question; but I don't have current access to AmberTools12, and I suspect others on the list are in the same situation.
...good luck....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Sep 09 2017 - 15:00:02 PDT
