AMBER Archive (2005)

Subject: Re: AMBER: ptraj analysis

From: David A. Case (case_at_scripps.edu)
Date: Wed Jan 19 2005 - 14:24:58 CST

On Wed, Jan 19, 2005, Thomas E. Cheatham, III wrote:
>
> matrix correl name correlation @CA out correl-matrix
> analyze matrix covariance out evectors name evecs vecs 255
>
> The first problem is about the number of eigenvectors and eigenvalues.
> My protein has 85 amino acids and using CA should produce 255
> eigenvectors and eigenvalues but in my output file (evectors) there are
> only 254 items and one of them is missing.

The principal component analysis is designed to be efficient at obtained
a small number of principal directions that contribute most to the observed
fluctuations. For technical reasons, it can compute at most N-1 vectors.
If you really need all of the vectors, you will need to modify the code.

...dac

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu