[phenixbb] Optimal settings for finding NCS from density

[Terwilliger, Thomas Charles]

Hi William,

Yes, that is a good point.  I'm not sure which way that goes though...you want the separation small so that adjacent molecules can be found, but large so that more of the long direction is explored.  I'd be inclined to go for a relatively small peak separation based on those considerations.

All the best,
Tom T

________________________________
From: William Chao [William.Chao at crick.ac.uk]
Sent: Sunday, July 05, 2015 12:42 PM
To: Terwilliger, Thomas Charles
Cc: phenixbb at phenix-online.org
Subject: Re: Optimal settings for finding NCS from density

Hi Tom,

Thanks very much for the tips. I wonder if I need to increase peak separation in case of a long protein?

Best,

William
---

From: "Terwilliger, Thomas Charles" <terwilliger at lanl.gov<mailto:terwilliger at lanl.gov>>
Date: Friday, 3 July 2015 18:08
To: William Chao <william.chao at crick.ac.uk<mailto:william.chao at crick.ac.uk>>
Cc: "phenixbb at phenix-online.org<mailto:phenixbb at phenix-online.org>" <phenixbb at phenix-online.org<mailto:phenixbb at phenix-online.org>>, "Terwilliger, Thomas Charles" <terwilliger at lanl.gov<mailto:terwilliger at lanl.gov>>
Subject: RE: Optimal settings for finding NCS from density

Hi William,

For find_ncs_from_density the shape of the molecule itself is not so important. What matters is if there is some local part of the molecule that is recognizable in both copies.  So I would try default parameters first.  I assume you did that and it didn't give you a useful answer...so then I would do these things:

1. try many locations :  n_center_use=20
This is the best option...it will try 20 times using density at 20 different locations to search.  If you have an older version of phenix you may need to download a nightly build to do this.

2. Try varying the resolution

3. Try varying the angular spacing of the search (delta_phi)

All the best,
Tom T

________________________________
From: CCP4 bulletin board [CCP4BB at JISCMAIL.AC.UK<mailto:CCP4BB at JISCMAIL.AC.UK>] on behalf of William Chao [William.Chao at CRICK.AC.UK<mailto:William.Chao at CRICK.AC.UK>]
Sent: Friday, July 03, 2015 10:24 AM
To: CCP4BB at JISCMAIL.AC.UK<mailto:CCP4BB at JISCMAIL.AC.UK>
Subject: [ccp4bb] Optimal settings for finding NCS from density

Dear all,

What would be the best settings to look for NCS using phenix.find_ncs_from_density on an elongated (~150A) molecule? No coordinate is available unfortunately and there should be 2 copies in a P21 space group.

Many thanks in advance!

William
---

The Francis Crick Institute Limited is a registered charity in England and Wales no. 1140062 and a company registered in England and Wales no. 06885462, with its registered office at 215 Euston Road, London NW1 2BE.

The Francis Crick Institute Limited is a registered charity in England and Wales no. 1140062 and a company registered in England and Wales no. 06885462, with its registered office at 215 Euston Road, London NW1 2BE.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20150705/9e5f7dc8/attachment.htm>