[sharp-discuss] very low B-factors
Clemens Vonrhein
vonrhein@globalphasing.com
Thu, 17 Jul 2003 08:27:02 +0100
--8nsIa27JVQLqB7/C
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
Dear Greg,
in my experience, B factors can go a bit funny at lowish resolution -
_if_ you also refine scaling B factors etc.=20
So could you please send us your whole SIN file? And have a look at
the values and the refinement behaviour of
SCAL_B
SCAL_B6_ADD
NISO_BGLO
NANO_BGLO
If SCAL_B parameters are going rather high (say > 5) I would probably
set them to zero and not refine/estimate them at all. Same for the
*_BGLO parameters.
And after your refinement (with or without fixed SCAL_B): what do the
residual maps look like? I know, it is nice if the values for
occupancy and B for sites come out at chemical meaningful
values. However, at this stage you are interested in getting a heavy
atom model that explains the differences (anomalous and dispersive)
well. So what I would do is
1. Have a look at the ANO residual maps for each wavelength will
tell you if the f'' values are correct. Since you use the same
HAT_OCC/HAT_B for both wavelengths (which makes sense) a wrong
f'' value at one wavelength can mess things up.
2. In the various residual maps, you would also see additional sites
(assuming that one of the 17 sites the one might be wrong).
3. Met sidechains quite often are disordered: any sign for that in
the residual maps? Also, N-terminal Met are often not visible.
4. Use autoSHARP (with the set of known sites) to run the whole
thing through to density modification. The difference in CC
between original and inverted hand will tell you if the HA model
is good. It will also give a whole lot of information (notes and
warnings - have a look at the warning messages!).
5. Why do you have a 2-wavelength MAD experiment? If this is because
your crystal slowly died (and you couldn't collect the third
wavelength) then I would tend to first do a SAD with the first
wavelength - and create a HA model that cleans up the residual
maps. Then add the second wavelength and see if it actually
helps.
6. Once you have done some density modification (and the CC between
the two hands is significantly different to be sure you have the
right hand) you can do 'ordinary' difference Fourier maps to get
a list of sites.
Hope that helps - let us know about any solution!
Cheers
Clemens
On Wed, Jul 16, 2003 at 05:42:35PM -0700, Gregory D. Bowman wrote:
> Hello -
>=20
> I am working with a 2-wavelength MAD dataset to about 3.2A resolution. Fr=
om a=20
> related crystal, I know the 17 Se sites pretty confidently, and these sit=
es=20
> seem to refine OK in SOLVE for the dataset I am having trouble with (so I=
am=20
> pretty sure the sites are all real). With SHARP, the B-factor of one site=
=20
> refines to 1.0. If I do not refine occupancies, the B-factor still dips p=
retty=20
> low in the first cycle (B=3D2.5) but manages to recover somewhat by the f=
inal=20
> cycle (B=3D56.7), though it is still relatively low for this resolution. =
Is it=20
> the case that I simply do not have enough information to refine these=20
> parameters at this resolution? Some of the logfile is below.
>=20
> Thanks for any advice.
>=20
> Greg
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D
> NOTE - site 6 is the offender. I apologise if the formatting does not car=
ry=20
> though in the email.
> -------------------------------------------------------
> refining both B and occupancies:
> --------------------------------
> First cycle of refining B/occupancies: Everything seems OK
>=20
>=20
> 001__G-SITE-001 HAT_OCC 1.1969 0.5500 1.7469
> HAT_B 149.2960 -1.6175 147.6785
> 001__G-SITE-002 HAT_OCC 1.5708 0.7393 2.3100
> HAT_B 155.8439 -1.5838 154.2601
> 001__G-SITE-003 HAT_OCC 1.5014 0.6834 2.1847
> HAT_B 186.3591 -1.2411 185.1180
> 001__G-SITE-004 HAT_OCC 1.4379 0.6495 2.0874
> HAT_B 199.2562 -1.3835 197.8728
> 001__G-SITE-005 HAT_OCC 1.8110 0.8303 2.6413
> HAT_B 213.9898 -0.5395 213.4502
>=20
> 001__G-SITE-006 HAT_OCC 1.6478 0.7564 2.4041
> HAT_B 155.2214 -1.8708 153.3506
>=20
> 001__G-SITE-007 HAT_OCC 1.1981 0.5572 1.7553
> HAT_B 174.7769 -1.0286 173.7482
> 001__G-SITE-008 HAT_OCC 1.3638 0.6294 1.9932
> HAT_B 185.8664 -0.9468 184.9196
> 001__G-SITE-009 HAT_OCC 1.6190 0.7405 2.3595
> HAT_B 193.9789 -1.0006 192.9783
> 001__G-SITE-010 HAT_OCC 1.4603 0.6834 2.1437
> HAT_B 200.6610 -0.5527 200.1083
> 001__G-SITE-011 HAT_OCC 1.4920 0.6880 2.1800
> HAT_B 168.2568 -1.9167 166.3402
> 001__G-SITE-012 HAT_OCC 1.4472 0.7089 2.1562
> HAT_B 225.7435 -0.9393 224.8042
> 001__G-SITE-013 HAT_OCC 0.9212 0.4453 1.3665
> HAT_B 163.8727 -0.9562 162.9165
> 001__G-SITE-014 HAT_OCC 1.1805 0.4887 1.6692
> HAT_B 264.7068 -0.5238 264.1831
> 001__G-SITE-015 HAT_OCC 1.3148 0.5880 1.9028
> HAT_B 250.8974 -1.0426 249.8548
> 001__G-SITE-016 HAT_OCC 1.4841 0.6934 2.1775
> HAT_B 300.0000 0.0000 300.0000 ->]
> 001__G-SITE-017 HAT_OCC 0.6966 0.3615 1.0581
> HAT_B 300.0000 0.0000 300.0000 ->]
> --------------------------------
> The very next small cycle: B-factor of SITE 6 drops by about 75
>=20
> 001__G-SITE-001 HAT_OCC 1.7469 0.0017 1.7486
> HAT_B 147.6785 -42.3724 105.3061
> 001__G-SITE-002 HAT_OCC 2.3100 -0.0154 2.2946
> HAT_B 154.2601 -31.3099 122.9502
> 001__G-SITE-003 HAT_OCC 2.1847 -0.0207 2.1640
> HAT_B 185.1180 8.1799 193.2979
> 001__G-SITE-004 HAT_OCC 2.0874 -0.0284 2.0590
> HAT_B 197.8728 19.6576 217.5304
> 001__G-SITE-005 HAT_OCC 2.6413 0.0608 2.7021
> HAT_B 213.4502 14.4088 227.8590
>=20
> 001__G-SITE-006 HAT_OCC 2.4041 0.0002 2.4044
> HAT_B 153.3506 -75.3947 77.9559
>=20
> 001__G-SITE-007 HAT_OCC 1.7553 -0.0193 1.7360
> HAT_B 173.7482 6.3321 180.0803
> 001__G-SITE-008 HAT_OCC 1.9932 0.0283 2.0215
> HAT_B 184.9196 18.8103 203.7298
> 001__G-SITE-009 HAT_OCC 2.3595 0.0342 2.3937
> HAT_B 192.9783 15.1212 208.0994
> 001__G-SITE-010 HAT_OCC 2.1437 -0.0026 2.1411
> HAT_B 200.1083 11.6774 211.7856
> 001__G-SITE-011 HAT_OCC 2.1800 -0.0634 2.1166
> HAT_B 166.3402 -56.0831 110.2571
> 001__G-SITE-012 HAT_OCC 2.1562 0.0179 2.1741
> HAT_B 224.8042 38.3077 263.1119
> 001__G-SITE-013 HAT_OCC 1.3665 -0.0235 1.3430
> HAT_B 162.9165 22.0442 184.9607
> 001__G-SITE-014 HAT_OCC 1.6692 0.0306 1.6998
> HAT_B 264.1831 9.4390 273.6221
> 001__G-SITE-015 HAT_OCC 1.9028 0.0364 1.9392
> HAT_B 249.8548 -4.5829 245.2718
> 001__G-SITE-016 HAT_OCC 2.1775 -0.0052 2.1722
> HAT_B 300.0000 0.0000 300.0000 <-]
> 001__G-SITE-017 HAT_OCC 1.0581 0.0441 1.1022
> HAT_B 300.0000 0.0000 300.0000 <-]
> -------------------------------------
> A few cycles later: B-factor of SITE 6 is now 1.0, and stays that way
>=20
> 001__G-SITE-001 HAT_OCC 0.9540 0.0608 1.0148
> HAT_B 100.8967 2.4084 103.3051
> 001__G-SITE-002 HAT_OCC 1.3670 0.0974 1.4644
> HAT_B 124.2807 -6.2762 118.0045
> 001__G-SITE-003 HAT_OCC 1.4122 0.0847 1.4969
> HAT_B 165.5046 -13.7893 151.7153
> 001__G-SITE-004 HAT_OCC 1.3042 0.1374 1.4417
> HAT_B 159.0778 14.0159 173.0936
> 001__G-SITE-005 HAT_OCC 1.8778 0.2253 2.1031
> HAT_B 227.7046 -4.6517 223.0529
>=20
> 001__G-SITE-006 HAT_OCC 1.0219 -0.2701 0.7518
> HAT_B 41.3661 -40.3661 1.0000 [<-
>=20
> 001__G-SITE-007 HAT_OCC 1.0922 0.0566 1.1488
> HAT_B 150.3008 -15.5740 134.7268
> 001__G-SITE-008 HAT_OCC 1.4094 0.1122 1.5216
> HAT_B 201.8194 -18.7466 183.0728
> 001__G-SITE-009 HAT_OCC 1.6166 0.2004 1.8171
> HAT_B 201.2297 -4.5830 196.6467
> 001__G-SITE-010 HAT_OCC 1.4988 0.1621 1.6608
> HAT_B 215.9288 -10.5880 205.3408
> 001__G-SITE-011 HAT_OCC 1.1856 0.0182 1.2038
> HAT_B 103.9965 5.4042 109.4008
> 001__G-SITE-012 HAT_OCC 0.8806 0.0912 0.9717
> HAT_B 55.0457 88.9120 143.9577
> 001__G-SITE-013 HAT_OCC 0.9247 0.1006 1.0253
> HAT_B 155.5759 6.6344 162.2103
> 001__G-SITE-014 HAT_OCC 1.1636 0.2149 1.3785
> HAT_B 287.3449 11.9467 299.2916
> 001__G-SITE-015 HAT_OCC 1.2764 0.0898 1.3662
> HAT_B 238.9843 -36.9708 202.0135
> 001__G-SITE-016 HAT_OCC 1.3258 0.1084 1.4342
> HAT_B 294.1810 -26.6462 267.5348
> 001__G-SITE-017 HAT_OCC 0.7161 0.1652 0.8813
> HAT_B 300.0000 0.0000 300.0000 <-]
> 001_001_001_001 NANO_BGLO 0.0000 0.0000 0.0000 [<-
>=20
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D
> not refining occupancies(occup=3D1.0)
> --------------------------------------------------------
> first cycle of B-factor refinement: B-factor of SITE 6 drops by 152
>=20
> 001__G-SITE-001 HAT_B 149.2960 -79.1425 70.1535
> 001__G-SITE-002 HAT_B 155.8439 -46.2592 109.5847
> 001__G-SITE-003 HAT_B 186.3591 -12.7966 173.5625
> 001__G-SITE-004 HAT_B 199.2562 -34.9708 164.2854
> 001__G-SITE-005 HAT_B 213.9898 -22.3762 191.6136
>=20
> 001__G-SITE-006 HAT_B 155.2214 -152.6520 2.5694
>=20
> 001__G-SITE-007 HAT_B 174.7769 -11.0692 163.7076
> 001__G-SITE-008 HAT_B 185.8664 -52.2808 133.5855
> 001__G-SITE-009 HAT_B 193.9789 -32.2719 161.7070
> 001__G-SITE-010 HAT_B 200.6610 -16.0723 184.5886
> 001__G-SITE-011 HAT_B 168.2568 -60.1715 108.0853
> 001__G-SITE-012 HAT_B 225.7435 -138.1051 87.6384
> 001__G-SITE-013 HAT_B 163.8727 -15.9228 147.9499
> 001__G-SITE-014 HAT_B 264.7068 1.3815 266.0884
> 001__G-SITE-015 HAT_B 250.8974 -25.9098 224.9876
> 001__G-SITE-016 HAT_B 300.0000 -6.1918 293.8082
> 001__G-SITE-017 HAT_B 300.0000 0.0000 300.0000 <-]
>=20
> ------------
>=20
> fourth cycle ("convergence")
> 001__G-SITE-001 HAT_B 92.8436 1.0171 93.8607 9.9228
>=20
> 001__G-SITE-002 HAT_B 119.0792 -0.3648 118.7144 8.7481
>=20
> 001__G-SITE-003 HAT_B 161.8942 5.2916 167.1858 11.5122
>=20
> 001__G-SITE-004 HAT_B 150.6342 3.9114 154.5456 11.5175
>=20
> 001__G-SITE-005 HAT_B 162.5293 4.9221 167.4514 10.9640
>=20
>=20
> 001__G-SITE-006 HAT_B 28.2706 12.8946 41.1652 2.8821
>=20
>=20
> 001__G-SITE-007 HAT_B 151.2232 4.0154 155.2387 13.1730
>=20
> 001__G-SITE-008 HAT_B 154.2849 3.8372 158.1221 11.7087
>=20
> 001__G-SITE-009 HAT_B 150.7050 2.1013 152.8063 11.0317
>=20
> 001__G-SITE-010 HAT_B 162.1460 6.4316 168.5776 12.3290
>=20
> 001__G-SITE-011 HAT_B 128.7785 3.6294 132.4079 8.5903
>=20
> 001__G-SITE-012 HAT_B 140.2975 21.5858 161.8834 10.1248
>=20
> 001__G-SITE-013 HAT_B 126.4959 -1.4780 125.0180 15.4597
>=20
> 001__G-SITE-014 HAT_B 223.7989 13.2720 237.0708 23.3402
>=20
> 001__G-SITE-015 HAT_B 154.6024 20.9541 175.5565 13.3123
>=20
> 001__G-SITE-016 HAT_B 236.7864 20.8837 257.6702 20.7670
>=20
> 001__G-SITE-017 HAT_B 300.0000 0.0000 300.0000 0.0000=
->]
>=20
> -------------
> last cycle (2 more small cycles)
>=20
> 001__G-SITE-001 HAT_B 87.9400 -5.2957 82.6443 12.1455
>=20
> 001__G-SITE-002 HAT_B 116.5185 -2.0488 114.4697 9.1728
>=20
> 001__G-SITE-003 HAT_B 167.9062 -0.4840 167.4222 11.8020
>=20
> 001__G-SITE-004 HAT_B 154.3282 -1.0146 153.3136 12.0092
>=20
> 001__G-SITE-005 HAT_B 168.5729 -0.4999 168.0730 11.3471
>=20
>=20
> 001__G-SITE-006 HAT_B 50.8219 5.8795 56.7014 5.8259
>=20
>=20
> 001__G-SITE-007 HAT_B 156.5339 -0.1544 156.3795 13.5625
>=20
> 001__G-SITE-008 HAT_B 158.7488 -0.4433 158.3056 12.1579
>=20
> 001__G-SITE-009 HAT_B 151.7882 -1.0714 150.7168 11.1549
>=20
> 001__G-SITE-010 HAT_B 169.1792 -0.2711 168.9081 12.8765
>=20
> 001__G-SITE-011 HAT_B 131.4322 -1.5125 129.9198 9.1421
>=20
> 001__G-SITE-012 HAT_B 167.3042 -0.7311 166.5730 13.2964
>=20
> 001__G-SITE-013 HAT_B 124.8798 -0.7616 124.1182 15.5111
>=20
> 001__G-SITE-014 HAT_B 238.3301 -0.2120 238.1182 25.2273
>=20
> 001__G-SITE-015 HAT_B 182.6898 1.4207 184.1104 16.7740
>=20
> 001__G-SITE-016 HAT_B 262.7511 -0.2747 262.4764 24.3344
>=20
> 001__G-SITE-017 HAT_B 300.0000 0.0000 300.0000 0.0000=
->]
>=20
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>=20
>=20
> _______________________________________________
> sharp-discuss mailing list
> sharp-discuss@globalphasing.com
> http://www.globalphasing.com/mailman/listinfo/sharp-discuss
--=20
***************************************************************
* Clemens Vonrhein, Ph.D. vonrhein AT GlobalPhasing DOT com
*
* Global Phasing Ltd.
* Sheraton House, Castle Park Tel: +44-(0)1223-353033
* Cambridge CB3 0AX, UK Fax: +44-(0)1223-366889
*--------------------------------------------------------------
* BUSTER Development Group (http://www.globalphasing.com)
***************************************************************
--8nsIa27JVQLqB7/C
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
iD8DBQE/Fk/G4f+nTPEF25QRApMQAJ4n1miILUWcnEL8U4ha5vJDkasHkgCgld0D
Qbk7FBQfpVAuD6bPuAqQtwk=
=xsx1
-----END PGP SIGNATURE-----
--8nsIa27JVQLqB7/C--