Dear Crystallographers,
in trying to convert a mmcif to mtz, I get the logfile below. I looked
in the directory, and there is a file cf_mm.dic, but this is
presumably not the same as the similar .lib file. Any thoughts about
this? Did the file somehow get lost? Also, I recently did this same
conve
Dear Colleague,
Due to many requests, we have extended the deadline for submission of abstracts
for contributed talks within:
International Conference
Multi-Pole Approach to Structural Biology
November 16-19, 2011 | Warsaw, Poland
Information, programme
Dear Ritcha,
we have crystallised various peptides, especially circular ones, both from
regular screens used in MX and from specially prepared screens (i.e. different
organic solvents).
We tend to use slow evaporation from under mineral oil in Terazaki plates (i.e.
microbatch).
We have also fa
On Tue, 2011-09-06 at 11:51 -0500, Chaudhary, Ritcha wrote:
> Do people use regular screens used in macromolecular crystallization
> such as Hamptons, wizards etc?
Probably not - I believe short peptides are crystallized mostly via
various evaporation techniques (but I am very curious to hear what
Dear all
I am interested in crystallizing a 5 residue peptide. I have no prior
experience in this field although I have crystallized larger proteins (5--60
kDa). Do people use regular screens used in macromolecular crystallization
such as Hamptons, wizards etc? Any suggestions are greatly app
On 06/09/11 16:48, Huw Jenkins wrote:
> On 25 Aug 2011, at 14:12, Gregory Bowman wrote:
>
>> When I try to run ARP/wARP classic for loop building, I get the following
>> message in the logfile:
>>
>> QUITTING ... ARP/wARP module stopped with an error message:
>> REFMAC
>
> I get the same error r
On 25 Aug 2011, at 14:12, Gregory Bowman wrote:
> When I try to run ARP/wARP classic for loop building, I get the following
> message in the logfile:
>
> QUITTING ... ARP/wARP module stopped with an error message:
> REFMAC
I get the same error running auto_tracing.sh from ARP/wARP 7.2. (CCP4 6
Dear SnowDeer,
Just how small is too small? If you have access to a microfocus beamline you
might find you can collect decent diffraction data from crystals with
dimensions in the single digit microns. Fishing tiny crystals is difficult,
but something like the MicroMesh "tennis racquet" mounts can
I have seen these type of cases. They cause problem. They may come from O, from
cns or other programs. Origins may even not be known. My solution was that if
an atom has coordinates 9998.000 or more then consider them absent (i.e. set
occupancies 0).
On 6 Sep 2011, at 14:30, Kevin Cowtan wr
OK, here's the solution (also to BB in case anyone else has this problem).
The PDB file appears to come from CNS.
The coordinate file contains the following atoms:
ATOM 2244 CG LEU A 452.000.000.000 1.00 0.00 A
ATOM 2245 CD1 LEU A 452.000.000.000 1
Can you send the command script, and mapdump output for any input maps?
Also, if you are entering a PDB file, the cell, spacegroup, and an idea
of range of X, Y, and Z values?
On 09/06/2011 09:56 AM, Ingo P. Korndoerfer wrote:
hello,
it seems i am recently getting the above error quite a lot
hello,
it seems i am recently getting the above error quite a lot in mapmask. i
have searched for a solution, but only found the problem mentioned a
few times with no precise solution indicated.
would there be a way to find out HOW MUCH memory mapmask is hoping for ?
i do have 2 Gb of memory and
Dear Petr,
Yes, please do send the log files, MTZ file and anything else needed to run the
job that shows the problem, so we can track it down. It's probably best to
send them off-line.
We couldn't reproduce Alexander's problem, but he tells me that it only occurs
"when the script is started
Dear all,
with the new MG version the program wants to start the previous session
giving me an empty window without any buttons no matter how hard I delete or
rename previous .CCP4MG files (presumably containing the status files
etc...).
As a result I cannot start new MG under my login ...
Any id
Dear Randy and the Phaser team,
Phaser 2.3.0 brought us several enhancement, but similar to Alexander
Schiffer's experience, Phaser has failed several times now with a floating
point exception error at different stages of automated MR: it has stopped in
the beginning of RF and then in a new run
15 matches
Mail list logo