[phenixbb] noisy sa-omit maps in solvent region.

Terwilliger, Thomas C terwilliger at lanl.gov
Thu Jan 19 13:17:27 PST 2012

Hi Francis,

I would expect your composite sa-omit maps to have significantly higher noise in the solvent region than a 2Fo-Fc map or a density modified map, because there is no solvent flattening applied and the map is (more or less) unbiased, so you see the actual noise level.

Autobuild simply stitches together the omit regions from each SA-omit run in real space.  No density modification is involved.  Map coefficients are back-calculated from the composite omit map and these are what you get from the SA-omit process.

All the best,
Tom T

From: phenixbb-bounces at phenix-online.org [phenixbb-bounces at phenix-online.org] on behalf of Francis E Reyes [Francis.Reyes at colorado.edu]
Sent: Thursday, January 19, 2012 2:11 PM
To: PHENIX user mailing list
Subject: [phenixbb] noisy sa-omit maps in solvent region.

Hi phenixers

I'm calculating a composite sa-omit map for my entire molecule at 4 A. While my molecule more or less has density (very good for backbone, not so much for sidechains), the solvent region seems to be particularly noisy.

Is this the normal experience for composite sa-omit maps at this resolution?

I'm particularly interested in what autobuild does when it 'stitches' the individual maps together, is it a direct stitch of the 2fo-fc's after refinement or is there some kind of real space density modification involved? Or a combination of both?

Any suggestions appreciated,


Francis E. Reyes M.Sc.
215 UCB
University of Colorado at Boulder

phenixbb mailing list
phenixbb at phenix-online.org

More information about the phenixbb mailing list