Hi,
On Thu, Jul 25, 2013 at 4:49 PM, Pavan G <pavanamber.gmail.com> wrote:
> Should I then consider the inverse of each value?
Not exactly. Eigenvalues are converted into frequencies with the
following formula (you can see this in the code for ptraj/cpptraj):
Freq = 108.587 * sqrt( 0.6 / Eigenvalue )
The 0.6 is kT at 300 K in kcal/mol and the 108.587 is a conversion
from Amber units to cm^-1. Hope this helps,
-Dan
> 5562 5562
> 1 1/0.80680
> 2 1/1.09819
> 3 1/1.67464
> 4 1/2.01560
> 5 1/2.39907
>
> Then the contribution of eigenvector
> 1 is 1/0.806 / [1/0.806 + 1/1.09 + ...+ 1/2.399]
> 4 is 1/2.015 / [1/0.806 + 1/1.09 + ...+ 1/2.399]
>
> This would make sense. Is this accurate?
>
> P
>
>
> On Thu, Jul 25, 2013 at 6:09 PM, Daniel Roe <daniel.r.roe.gmail.com> wrote:
>
>> Hi,
>>
>> On Thu, Jul 25, 2013 at 2:20 PM, Pavan G <pavanamber.gmail.com> wrote:
>> > But this would mean that contribution of 2nd eigenvector is more than 1st
>> > eigenvector which seems incorrect since
>> > 1>2>3>4>5.
>>
>> When you analyze mass-weighted covariance matrices (quasi-harmonic
>> analysis) the eigenvalues are converted into units of cm^-1, so the
>> smallest # in your case does correspond to the largest eigenvalue.
>>
>> -Dan
>>
>> --
>> -------------------------
>> Daniel R. Roe, PhD
>> Department of Medicinal Chemistry
>> University of Utah
>> 30 South 2000 East, Room 201
>> Salt Lake City, UT 84112-5820
>> http://home.chpc.utah.edu/~cheatham/
>> (801) 587-9652
>> (801) 585-9119 (Fax)
>>
>> _______________________________________________
>> 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
--
-------------------------
Daniel R. Roe, PhD
Department of Medicinal Chemistry
University of Utah
30 South 2000 East, Room 201
Salt Lake City, UT 84112-5820
http://home.chpc.utah.edu/~cheatham/
(801) 587-9652
(801) 585-9119 (Fax)
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Jul 25 2013 - 22:30:04 PDT
