[sharp-discuss] H32 SHARP-ARP/wARP-Refmac problem?
Clemens Vonrhein
vonrhein@globalphasing.com
Tue, 10 Dec 2002 11:37:13 +0000
Dear Ole,
please note, that the current version of SHARP/autoSHARP (3.0.14) is
not tested yet under CCP4 4.2.2! I'm working on porting CCP4 4.2.2 to
my set of patches (see
http://www.globalphasing.com/people/vonrhein/ccp4/) and then get a new
version of SHARP/autoSHARP out that will be tested with 4.2.2.
However, there might be a problem/feature with H32 - since this is a
fairly new change to several programs (previously, it was called R32
even if hexagonal setting). But if this was a problem I would think
that REFMAC should stop much earlier. So my guess is, that the refmac
version you're running has a problem. Did you get in contact with
Garib about that?
Also, you might want to use the standalone ARP/wARP 6.0
version. Although I tried to optimize the interface to the various
ARP/wARP 6.0 programs to perform optimally with the SHARP output,
there might still be a problem.
Hope that helps.
Cheers
Clemens
On Tue, Dec 10, 2002 at 11:22:04AM +0100, Ole Kristensen wrote:
> In a current case, I got so far as to run the arp/warp step from autosharp
> (SHARP 2.0.0, Sushi 3.0.14). Despite good phases I got stuck. Here is the
> log output:
> MTZ file :
> /home/sharp/users/olek/None.sharp/logfiles/tig118.5/eden_flat_34.5pc.mtz
>
> $warpbin : /a/xtal/linux/arp_warp_6.0/bin/Linux
> Cell : 71.6440 71.6440 375.4090 90.0000 90.0000 120.0000
> Spacegroup : H32 (155)
> Wilson B-Factor : 38.419
> Protein size : 2952
> Sequence file : <wARP_34.5pc/seq.pir>
> Resolution : 30.000 2.30
> Asymmetric unit : 0.00000 1.00000 0.00000 1.00000 0.00000 1.00000
> 2. Generating "warp.par" for ARP/wARP version 6.0
> (details) <wARP_34.5pc/warp.par> (directory index) <wARP_34.5pc>
> 3. Running "arpwarp.sh"
> 1. reading warp.par (for ARP/wARP version 6.0)
> 2. Building free atoms model
> 2.1 calculating initial map
> 2.2 expanding map to ARP limits
> 2.3 building initial model
> Note: Had to go as low as 0.80 sigma, to complete atoms search.
>
> Note: Remove atoms if below 1.0 sigmas in 2mFoDFc map.
> Add atoms in density above 3.2 sigmas in mFoDFc map.
> A maximum of 100 ARP refinement cycles will be run in total.
> Rebuilding of the model will happen every 10 cycles.
> 3. running warpNtrace loop
>
> Note: using fixed weight in REFMAC of 0.5 - check geometry afterwards!
>
>
> Note: automatically adjusting phase blurring in REFMAC
>
> 16247 reflections ( 99.99 % complete ) and 0 restraint used, for
> refining 3310 atoms.
> Observations/parameters ratio is 1.23 (~1.23 with restraints).
>
> Starting model : R = 0.341 (Rfree = 0.331)
>
> Cycle 1 : After refmac : R = 0.314 (Rfree = 0.320)
> Found 21 (21 requested) and removed 15 (16 requested) atoms.
> Cycle 2 : After refmac : R = 0.297 (Rfree = 0.321)
> Found 21 (21 requested) and removed 2 (16 requested) atoms.
> Cycle 3 : After refmac : R = 0.286 (Rfree = 0.324)
> Found 21 (21 requested) and removed 5 (18 requested) atoms.
> Cycle 4 : After refmac : R = 0.278 (Rfree = 0.322)
> Found 22 (22 requested) and removed 9 (21 requested) atoms.
> Cycle 5 : After refmac : R = 0.271 (Rfree = 0.321)
> Found 22 (22 requested) and removed 10 (22 requested) atoms.
> Cycle 6 : After refmac : R = 0.265 (Rfree = 0.317)
> Found 22 (22 requested) and removed 13 (22 requested) atoms.
> Cycle 7 : After refmac : R = 0.260 (Rfree = 0.315)
> Found 22 (22 requested) and removed 20 (22 requested) atoms.
> Cycle 8 : After refmac : R = 0.255 (Rfree = 0.312)
> Found 22 (22 requested) and removed 12 (22 requested) atoms.
> Cycle 9 : After refmac : R = 0.253 (Rfree = 0.313)
> Found 22 (22 requested) and removed 21 (22 requested) atoms.
> Cycle 10 : After refmac : R = 0.249 (Rfree = 0.309)
> Found 22 (22 requested) and removed 7 (22 requested) atoms.
>
> Building Cycle 1
> Round 1 : 291 peptides in 22 chains.
> Round 2 : 312 peptides in 14 chains.
> Chains 14, Residues 298 [out of 369], Connectivity Index 0.91
>
> Cycle 11 : After refmac : R = 0.339 (Rfree = 0.384)
> Found 16 (16 requested) and removed 8 (8 requested) atoms.
> Cycle 12 : After refmac : R = 0.317 (Rfree = 0.375)
> Found 16 (16 requested) and removed 8 (8 requested) atoms.
> Cycle 13 : After refmac : R = 0.303 (Rfree = 0.366)
> Found 16 (16 requested) and removed 8 (8 requested) atoms.
> Cycle 14 : After refmac : R = 0.293 (Rfree = 0.359)
> Found 16 (16 requested) and removed 8 (8 requested) atoms.
> Cycle 15 : After refmac : R = 0.283 (Rfree = 0.348)
> Found 16 (16 requested) and removed 8 (8 requested) atoms.
> Cycle 16 : After refmac : R = 0.277 (Rfree = 0.345)
> Found 16 (16 requested) and removed 8 (8 requested) atoms.
> Cycle 17 : After refmac : R = 0.270 (Rfree = 0.323)
> Found 17 (17 requested) and removed 8 (8 requested) atoms.
> Cycle 18 : After refmac : R = 0.263 (Rfree = 0.321)
> Found 17 (17 requested) and removed 8 (8 requested) atoms.
> Cycle 19 : After refmac : R = 0.258 (Rfree = 0.317)
> Found 17 (17 requested) and removed 8 (8 requested) atoms.
> Cycle 20 : After refmac : R = 0.251 (Rfree = 0.316)
> Found 17 (17 requested) and removed 8 (8 requested) atoms.
>
> Building Cycle 2
> Round 1 : 285 peptides in 14 chains.
> Round 2 : 307 peptides in 13 chains.
> Chains 10, Residues 285 [out of 369], Connectivity Index 0.92
>
> Note: had to fix PDB file "./files/reshuffle_cycle20.pdb" for RES. numbering
> - please check programs!
>
> ERROR : program terminated in REFMAC5 (see refmac_cycle21.log
> <wARP_34.5pc/refmac_cycle21.log>)
>
> The ./files/reshuffle_cycle20.pdb looks okay. Refmac seems to be the
> problem. I tried
> both the refmac5 version from ccp4-4.2.2 and garib's latest version. Here,
> is a bit
> of my latest attempt:
> ###############################################################
> ###############################################################
> ###############################################################
> ### CCP4 4.2: Refmac_5.2.00 version 5.2.00 : 31/10/01##
> ###############################################################
> User: Run date: 9/12/02 Run time:23:43:19
> ......
> Data line--- MAKE HYDR N
> Data line--- MAKE CHECK 0
> Data line--- LABI FP=FPsha SIGFP=SIGFPsha FREE=FreeR_flag HLA=HLAdm
> HLB=HLBdm HLC=HLCdm HLD=HLDdm
> Data line--- LABO FC=FC PHIC=PHIC FWT=2FOFCWT PHWT=PH2FOFCWT
> DELFWT=FOFCWT PHDELWT=PHFOFCWT
> Data line--- REFI RESOLUTION 30.000 2.30 TYPE RESTR
> Data line--- REFI RESI MLKF PHASed SCBLurred 0.34
> Data line--- REFI BREF ISOTROPIC METH CGMAT
> Data line--- WEIG MATRIX 0.5
> Data line--- DAMP 1.0 1.0
> Data line--- SCAL TYPE BULK LSSC ANIS FIXBulk
> Data line--- SCAL MLSC FIXBulk
> Comment line--- ! maybe only at later stages?
> Data line--- SOLVENT YES
> Data line--- NCYC 1
> Data line--- MONI MANY
> Data line--- BLIMI 2 80
> Data line--- BINS 20
> Data line--- VDWR SIGMA DUMMY 50.0 INCREMENT DUMMY -1.0
> Data line--- VDWR 0.0
> Data line--- END
> .....
> .....
> CGMAT cycle number = 1
> Too many atoms in one brick.
> Maximum allowed in one brick = 60
> Check symmetry, cell dimensions
> ===> Error: CREATE_NET failed
> Refmac_5.2.00: CREATE_NET failed
> Times: User: 5.0s System: 0.7s Elapsed: 0:13
>
> Any ideas about how to get around this? I was wondering if it is related to
> the asymmetric unit being defined as:
> Asymmetric unit : 0.00000 1.00000 0.00000 1.00000 0.00000 1.00000
>
> Clemens has previously pointed out that one could possibly change the
> settings in
> $BDG_home/database/lists/symm.dat. Will it work to change the field 10 for
> spacegroup
> H32 to something smaller than the 0 1 0 1 0 1 limits distributed?
> When I simply take the starting file eden_flat_34.5pc.mtz and run arpwarp
> just from the
> ccp4i with default settings a smaller asu seems to be used and thing run
> without error.
>
>
> Best regards,
> Ole Kristensen
> Denmark
>
>
>
>
>
>
> _______________________________________________
> sharp-discuss mailing list
> sharp-discuss@globalphasing.com
> http://www.globalphasing.com/mailman/listinfo/sharp-discuss
--
***************************************************************
* 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)
***************************************************************