Hi Dmytro,

Thank you for your code snippet, I have a question in the code you sent.

In that you say:

BC = 1-det_XY/math.sqrt(det_XX_all[i]*det_XX_all[j])

Should it not be

BC = det_XY/math.sqrt(det_XX_all[i]*det_XX_all[j])

There should be no 1-det_XY as the numerator but instead det_XY?

Or have I made a mistake?

Thank you,

Patrick

-----Original Message-----
From: [email protected]
Sent: Thu, 03 Jul 2014 13:09:08 +0200
To: [email protected]
Subject: Re: [phenixbb] alternatives to RMSD
Hi Patrick

Take a look at the Binet-Cauchy kernel, e.g. here: http://bioinformatics.oxfordjournals.org/content/30/6/784.long

It is a shape-correlation metric that is rather insensitive to global shape changes, but sensitive to local ones. Maybe that's what you need. It's also much faster to compute than RMSD.

I have attached a python code snippet that computes a BC distance matrix for a set of proteins defined by CA coordinates.

Kind regards,
Dmytro.

On 02/07/14 17:15, Patrick. C wrote:
Hi Phenix users,

I am not a crystallographer but I though you guys might be a good place to ask this question.

I have 2 super secondary structures, A and B and they consist of Helix-turn-Strand

Due to the turn the two structures have a poor RMSD because the two flanking fragments of Helix and Strand are far from each other but when I superimpose the two fragments individually(helixA with helix B and standA with strandB in Pymol they align very well).

Now, is there a way to express this instead of using the RMSD?

When the two structures align well the RMSD is very good but a slight movement and the RMSD is awful.

But looking at the two structures I can see they follow the same path through space.

Thank you,

Patrick

Free 3D Earth Screensaver
Watch the Earth right on your desktop! Check it out at www.inbox.com/earth
_______________________________________________
phenixbb mailing list
[email protected]
http://phenix-online.org/mailman/listinfo/phenixbb

3D Earth Screensaver Preview

Free 3D Earth Screensaver
Watch the Earth right on your desktop! Check it out at www.inbox.com/earth