AMBER Archive (2008)

Subject: Re: AMBER: matrix correl

From: Holger Gohlke (gohlke_at_bioinformatik.uni-frankfurt.de)
Date: Mon Feb 18 2008 - 06:56:07 CST

Dear David,

the correlation matrix output on a per-residue basis is the _average_ over
correlations per atom contained in this residue, not the correlation of, say,
the movements of residue centers. As not all atoms within one residue move
perfectly correlated with respect to each other, the diagonal elements won't
be 1 either.

Best regards

Holger

Am Samstag, 16. Februar 2008 10:09 schrieb David Beveridge:
> Did this problem ever get diagnosed?
>
> DLB
>
> Dear Amber users,
>
> Does anyone have experience with correlation facility in Amber8? The
> residue-residue correlation map of "matrix" command has diagonal elements
> are not 1! It's weird because I thought the residue will move correlated
> with itself during the simulation; thus, the correlation should be 1.
>
> Here is my input file for ptraj:
>
> trajin corr500.crd.gz
> center :1-534 mass origin
> image origin center
> rms first mass out rmsh.dat :1-534
> matrix correl name corr :1-534 out dccm byres mass
>
> And here is a part of the output file (534x534 matrix, I took only 10x10
> matrix):
>
> 0.89 0.66 0.37 0.29 0.13 -0.03 -0.24 -0.27 -0.20 -0.08
> 0.66 0.89 0.58 0.42 0.25 0.03 -0.07 -0.19 -0.12 -0.08
> 0.37 0.58 0.75 0.66 0.47 0.25 0.24 0.07 0.14 0.12
> 0.29 0.42 0.66 0.90 0.73 0.47 0.40 0.23 0.32 0.32
> 0.13 0.25 0.47 0.73 0.85 0.60 0.53 0.38 0.44 0.39
> -0.03 0.03 0.25 0.47 0.60 0.77 0.65 0.61 0.57 0.56
> -0.24 -0.07 0.24 0.40 0.53 0.65 0.90 0.71 0.59 0.46
> -0.27 -0.19 0.07 0.23 0.38 0.61 0.71 0.90 0.77 0.68
> -0.20 -0.12 0.14 0.32 0.44 0.57 0.59 0.77 0.93 0.80
> -0.08 -0.08 0.12 0.32 0.39 0.56 0.46 0.68 0.80 0.85
>
> Any hints?
>
> Best regards,
> Bimo
>
>
> David L. Beveridge
> University Professor
> Chemistry Department and
> Molecular Biophysics Program
> Wesleyan University
> Middletown, CT 06459
> Ph 860 346 6768
> Fx 860 346 2211
>
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu 

-- 
++++++++++++++++++++++++++++++++++++++++++++++++++
Dr. Holger Gohlke
Juniorprofessor fuer Molekulare Bioinformatik

J.W. Goethe-Universitaet
Fachbereich Biowissenschaften
Institut fuer Zellbiologie und Neurowissenschaft
Max-von-Laue-Str. 9
60438 Frankfurt/Main
Germany

Tel.:  (+49) 69-798-29411; Fax: (+49) 69-798-29527
Email: gohlke_at_bioinformatik.uni-frankfurt.de
URL:   http://mbilab.uni-frankfurt.de
++++++++++++++++++++++++++++++++++++++++++++++++++
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu