Re: AMBER: ptraj analyze matrix entropy

From: David A. Case <case.scripps.edu>
Date: Thu, 6 May 2004 08:29:30 -0700

On Thu, May 06, 2004, Marcin Krol wrote:

> I have tried to calculate mass weighted covariance matrix and analyze it,
> obtaining entropy contributions. the ptraj script that does it is very
> simple:
>
> trajin traje.file (I have oriented all frames on the starting structure)
> matrix mwcowar name mwc200 out mwc200.dat start 1 stop 201
> analyze matrix mwc200 out mwc200_ana.dat thermo vecs 0
>
> However, if the number 3*N (N-atoms) is larger than 200 I get starnge
> results:
> 1. I get a warning: setting vibrational entropy to zero for mode 201 with
> vtemp = -319766250.735258

If you have only 200 snapshots, then the maximum number of quasiharmonic
modes you can obtain is 200 (independent of the number of atoms). This is
because the linear space spanned by your snapshots can be completely described
with 200 vectors, and no other ones make sense.

The program should give a better message in this case, but you can ignore
the results for modes that are greater than the number of snapshots.

However, you should also know that with so few snapshots, you will almost
certainly not get a converged estimate of the conformational entropy.

[Holger: we (you?) should probably adjust the code to recognize than all
modes above the number of snapshots used is irrelevant. But I'm not sure
how easy this will be, since the "analyze matrix" command may not necessarily
know how the matrix was constucted.]

....good luck...dac

-- 
==================================================================
David A. Case                     |  e-mail:      case.scripps.edu
Dept. of Molecular Biology, TPC15 |  fax:          +1-858-784-8896
The Scripps Research Institute    |  phone:        +1-858-784-9768
10550 N. Torrey Pines Rd.         |  home page:                   
La Jolla CA 92037  USA            |    http://www.scripps.edu/case
==================================================================
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Thu May 06 2004 - 16:53:00 PDT
Custom Search