b-factor restraints file

Gerard 'CD' Kleywegt GERARD at XRAY.BMC.UU.SE
Tue Jun 4 20:23:55 EST 1996


From:	SMTP%"msweiss at imb-jena.de" 13-MAY-1996 16:05:43.87
To:	"bionet.software.x-plor mail newsgroup" <bionet-news at dl.ac.uk>
CC:	
Subj:	Temperature factor restraints

Resent-From: <server-daemon at dl.ac.uk>
Date: 13 May 1996 06:01:52 -0700
Resent-Date: Mon, 13 May 96 14:22:22 UT
Message-ID: <96513142222.~INN-PZEa14669.bionet-news at dl.ac.uk>
Resent-Message-ID: <199605131255.OAA04169 at duncan.imb-jena.de>
Precedence: list
From: msweiss at imb-jena.de (Manfred Weiss)
Reply-To: msweiss at imb-jena.de (Manfred Weiss)
Sender: "bionet.software.x-plor mail newsgroup" <server-daemon at dl.ac.uk>
To: "bionet.software.x-plor mail newsgroup" <bionet-news at dl.ac.uk>
Subject: Temperature factor restraints
Comments: List problems/queries to <biosci at dl.ac.uk>
Comments: To mail both the group and netnews send to (x-plor at dl.ac.uk)
X-Article-Number: bionet.software.x-plor Msg # 511
X-Listpath: bionet-news
X-Mailer: MXT V 12.16.1

Dear Xplorers,

today I would like to start a discussion on temparature factor restraints.
Dale Tronrud published a study on knowledge based B-factor restraints for
the refinement of proteins recently (J. Appl. Cryst. 29, 100-4 (1996)). I
think the idea is a good one, although it was insufficiently tested (that's 
my own opinion of course). 

In brief, he derived B-factor restraints from a small sample of structures 
refined at very high resolution without restraints on the temperature 
factors. Then, a systematic difference between B-factors of bonded atoms as 
well as a standard deviation was computed. To test this briefly using one 
of our own in-house projects I used the standard deviations as target sigmas 
(bsigma in XPLOR), basically neglected all asigma values, and also neglected 
the systematic difference in Bs from bonded atoms. (For those who also want 
to do some testing, I have enclosed my input script for temperature factor 
refinement; saves some typing time). 

To summarize my test: 
1) Use standard XPLOR defaults for bsigmas (1.5 and 2.0) for main and 
side chain atoms and asigmas (2.0 and 2.5) for main and side chain atoms. 
2) Use defaults for bsigmas (1.5 and 2.5) for main and side chain atoms 
and asigmas (2.5 and 5.0) for main and side chain atoms.
3) Use Tronrud's target standard deviations (see appendix) for bonded
atoms, not standard deviations for atoms separated by two bonds.

Run 25 cycles of b-factor refinement, test all weights from 0.0 - 1.0
in 0.1 steps.

Results: whereas 1) and 2) hardly differ (assuming the same weighting
factor), 3) gives constantly better Rfree values (by the way, the
resolution of the data is 2.4 A). The improvement is about 0.7 - 1.0%
in Rfree. 

My questions:

- comments and criticism?
- do we need 'new' temperature factor restraints?
- does anybody have a different set in personal use that could
  be tested?
 
   Manfred Weiss.

 --
   Manfred S. Weiss, Dr.
   Institute of Molecular Biotechnology
   Department of Structural Biology and Crystallography
   Beutenbergstrasse 11         
   Postfach 100813                                   Phone: +49-3641-65-6070
   D-07708 Jena                                      Fax:   +49-3641-65-6062
   GERMANY                                           Email: msweiss at imb-jena.de

---------------

REMARKS file bref_tnt.restr
REMARKS contains Dale Tronrud's parameters for atomic temperature factor
REMARKS refinement (published in J. Appl. Cryst. 29, 100-104 (1996) )

REMARKS for main chain atoms, 898 examples

 bsigma = ( name N  or name CA ) = 4.6
 bsigma = ( name CA or name C  ) = 7.4
 bsigma = ( name C  or name N  ) = 6.6
 bsigma = ( name C  or name O  or name OT* ) = 6.3

REMARKS alanine side chains, 80 examples

 bsigma = ( resnam ALA and ( name CA or name CB ) ) = 3.5

REMARKS arginine side chains, 36 examples

 bsigma = ( resnam ARG and ( name CA or name CB ) ) = 8.5
 bsigma = ( resnam ARG and ( name CB or name CG ) ) = 16.5
 bsigma = ( resnam ARG and ( name CG or name CD ) ) = 21.3
 bsigma = ( resnam ARG and ( name CD or name NE ) ) = 22.7
 bsigma = ( resnam ARG and ( name NE or name CZ ) ) = 21.3
 bsigma = ( resnam ARG and ( name CZ or name NH1 ) ) = 23.5
 bsigma = ( resnam ARG and ( name CZ or name NH2 ) ) = 23.4
 
REMARKS asparagine side chains, 51(50) examples

 bsigma = ( resnam ASN and ( name CA or name CB ) ) = 10.8
 bsigma = ( resnam ASN and ( name CB or name CG ) ) = 25.0
 bsigma = ( resnam ASN and ( name CG or name OD1 ) ) = 21.1
 bsigma = ( resnam ASN and ( name CG or name ND2 ) ) = 18.9

REMARKS aspartate side chains, 55 examples

 bsigma = ( resnam ASP and ( name CA or name CB ) ) = 4.5
 bsigma = ( resnam ASP and ( name CB or name CG ) ) = 14.3
 bsigma = ( resnam ASP and ( name CG or name OD1 ) ) = 9.7
 bsigma = ( resnam ASP and ( name CG or name OD2 ) ) = 14.3

REMARKS cysteine side chains, 16 examples

 bsigma = ( resnam CYS and ( name CA or name CB ) ) = 3.8
 bsigma = ( resnam CYS and ( name CB or name SG ) ) = 3.2

REMARKS glutamine side chains, 37 examples

 bsigma = ( resnam GLN and ( name CA or name CB ) ) = 4.1
 bsigma = ( resnam GLN and ( name CB or name CG ) ) = 22.8
 bsigma = ( resnam GLN and ( name CG or name CD ) ) = 24.1
 bsigma = ( resnam GLN and ( name CD or name OE1 ) ) = 18.4
 bsigma = ( resnam GLN and ( name CD or name NE2 ) ) = 22.3

REMARKS glutamate side chains, 27 examples

 bsigma = ( resnam GLU and ( name CA or name CB ) ) = 3.5
 bsigma = ( resnam GLU and ( name CB or name CG ) ) = 13.5
 bsigma = ( resnam GLU and ( name CG or name CD ) ) = 25.1
 bsigma = ( resnam GLU and ( name CD or name OE1 ) ) = 33.9
 bsigma = ( resnam GLU and ( name CD or name OE2 ) ) = 26.9

REMARKS histidine side chains, 16 examples

 bsigma = ( resnam HIS and ( name CA or name CB ) ) = 3.8
 bsigma = ( resnam HIS and ( name CB or name CG ) ) = 3.8
 bsigma = ( resnam HIS and ( name CG or name ND1 ) ) = 4.0
 bsigma = ( resnam HIS and ( name CG or name CD2 ) ) = 5.4
 bsigma = ( resnam HIS and ( name ND1 or name CE1 ) ) = 4.3
 bsigma = ( resnam HIS and ( name CE1 or name NE2 ) ) = 3.3
 bsigma = ( resnam HIS and ( name CD2 or name NE2 ) ) = 3.7

REMARKS isoleucine side chains, 52 examples

 bsigma = ( resnam ILE and ( name CA or name CB ) ) = 4.2
 bsigma = ( resnam ILE and ( name CB or name CG1 ) ) = 5.0
 bsigma = ( resnam ILE and ( name CG1 or name CD1 ) ) = 17.2
 bsigma = ( resnam ILE and ( name CB or name CG2 ) ) = 6.9

REMARKS leucine side chains, 56 examples

 bsigma = ( resnam LEU and ( name CA or name CB ) ) = 3.6
 bsigma = ( resnam LEU and ( name CB or name CG ) ) = 5.3
 bsigma = ( resnam LEU and ( name CG or name CD1 ) ) = 5.6
 bsigma = ( resnam LEU and ( name CG or name CD2 ) ) = 9.2

REMARKS lysine side chains, 57 examples

 bsigma = ( resnam LYS and ( name CA or name CB ) ) = 5.8
 bsigma = ( resnam LYS and ( name CB or name CG ) ) = 21.2
 bsigma = ( resnam LYS and ( name CG or name CD ) ) = 18.9
 bsigma = ( resnam LYS and ( name CD or name CE ) ) = 24.1
 bsigma = ( resnam LYS and ( name CE or name NZ ) ) = 24.2

REMARKS methionine side chains, 12 examples

 bsigma = ( resnam MET and ( name CA or name CB ) ) = 3.4
 bsigma = ( resnam MET and ( name CB or name CG ) ) = 4.3
 bsigma = ( resnam MET and ( name CG or name SD ) ) = 4.7
 bsigma = ( resnam MET and ( name SD or name CE ) ) = 4.5

REMARKS phenylalanine side chains, 24 examples

 bsigma = ( resnam PHE and ( name CA or name CB ) ) = 3.8
 bsigma = ( resnam PHE and ( name CB or name CG ) ) = 4.5
 bsigma = ( resnam PHE and ( name CG or name CD1 ) ) = 6.2
 bsigma = ( resnam PHE and ( name CG or name CD2 ) ) = 6.2
 bsigma = ( resnam PHE and ( name CD1 or name CE1 ) ) = 5.3
 bsigma = ( resnam PHE and ( name CD2 or name CE2 ) ) = 5.3
 bsigma = ( resnam PHE and ( name CE1 or name CZ ) ) = 6.7
 bsigma = ( resnam PHE and ( name CE2 or name CZ ) ) = 5.1

REMARKS proline side chains, 25 examples

 bsigma = ( resnam PRO and ( name CA or name CB ) ) = 3.9
 bsigma = ( resnam PRO and ( name CB or name CG ) ) = 5.8
 bsigma = ( resnam PRO and ( name CG or name CD ) ) = 4.9
 bsigma = ( resnam PRO and ( name CD or name N ) ) = 5.6

REMARKS serine side chains, 68 examples

 bsigma = ( resnam SER and ( name CA or name CB ) ) = 7.1
 bsigma = ( resnam SER and ( name CB or name OG ) ) = 13.9

REMARKS threonine side chains, 72 examples

 bsigma = ( resnam THR and ( name CA or name CB ) ) = 13.6
 bsigma = ( resnam THR and ( name CB or name OG1 ) ) = 14.7
 bsigma = ( resnam THR and ( name CB or name CG2 ) ) = 12.6

REMARKS tryptophane side chains, 17 examples

 bsigma = ( resnam TRP and ( name CA or name CB ) ) = 3.8
 bsigma = ( resnam TRP and ( name CB or name CG ) ) = 3.5
 bsigma = ( resnam TRP and ( name CG or name CD1 ) ) = 2.8
 bsigma = ( resnam TRP and ( name CD1 or name NE1 ) ) = 4.2
 bsigma = ( resnam TRP and ( name NE1 or name CE2 ) ) = 4.3
 bsigma = ( resnam TRP and ( name CE2 or name CZ2 ) ) = 4.2
 bsigma = ( resnam TRP and ( name CZ2 or name CH2 ) ) = 6.0
 bsigma = ( resnam TRP and ( name CZ3 or name CH2 ) ) = 3.0
 bsigma = ( resnam TRP and ( name CE3 or name CZ3 ) ) = 3.8
 bsigma = ( resnam TRP and ( name CD2 or name CE3 ) ) = 3.4
 bsigma = ( resnam TRP and ( name CG or name CD2 ) ) = 3.0
 bsigma = ( resnam TRP and ( name CD2 or name CE2 ) ) = 5.3

REMARKS tyrosine side chains, 48 examples

 bsigma = ( resnam TYR and ( name CA or name CB ) ) = 4.7
 bsigma = ( resnam TYR and ( name CB or name CG ) ) = 6.0
 bsigma = ( resnam TYR and ( name CG or name CD1 ) ) = 4.8
 bsigma = ( resnam TYR and ( name CG or name CD2 ) ) = 4.6
 bsigma = ( resnam TYR and ( name CD1 or name CE1 ) ) = 7.9
 bsigma = ( resnam TYR and ( name CD2 or name CE2 ) ) = 4.4
 bsigma = ( resnam TYR and ( name CE1 or name CZ ) ) = 14.6
 bsigma = ( resnam TYR and ( name CE2 or name CZ ) ) = 12.2
 bsigma = ( resnam TYR and ( name CZ or name OH ) ) = 13.1

REMARKS valine side chains, 66 examples

 bsigma = ( resnam VAL and ( name CA or name CB ) ) = 4.8
 bsigma = ( resnam VAL and ( name CB or name CG1 ) ) = 4.3
 bsigma = ( resnam VAL and ( name CB or name CG2 ) ) = 3.8

REMARKS disulfide bridges, 9 examples

 bsigma = ( name SG ) = 2.1 
 
REMARKS no weights on angles

 asigma = ( not name H* ) = 1000000








More information about the X-plor mailing list