[phenixbb] MTZ intensities interpreted as amplitudes by autobuild command-line versions up to 1.3b/rc6

[Thomas C. Terwilliger]

Hi again Peter,

Yes, the AutoMR wizard (any version) will run correctly with input MTZ
data that are intensities.  However all versions up to now will pass this
MTZ file on to AutoBuild without conversion to amplitudes and it will be
incorrectly interpreted by AutoBuild. Note that this only applies to MTZ
files. Any other source of amplitudes/intensities doesn't cause any
problem.

If you run from the GUI, you will have the chance to set the correct
columns in AutoBuild, if amplitudes are available in your data file.  From
the command_line, for existing versions, you would need to say when you
run AutoMR:

input_labels='F SIGF'  (or whatever your amplitudes are in this MTZ file)

All the best,
Tom T



> Hi Tom,
>
> What about the AutoMR wizard. Can I use intensities there instead of
amplitudes? I always get confused.
>
> Thanks,
>
> Peter
>
> 2007/12/18, Tom Terwilliger <terwilliger at lanl.gov>:
>> Hi phenix.autobuild users,
>> I'm sorry to report a serious bug that I have just noticed that applies
to MTZ files for autobuild containing intensities. The
>> problem is fixed now. I'm very sorry about this mistake.
>> All the best,
>> -Tom T
>>
>>
>> THE PROBLEM:
>>
>> In all alpha and beta versions of phenix up to now (up to 1.3b/rc6),
there has been a bug in autobuild in which the command
>>
>> phenix.autobuild data.mtz model.pdb
>>
>> will misinterpret intensity data as amplitudes if the input file is an
mtz file (for example if the mtz file has IMEAN,SIGIMEAN and F,SIGF). 
(This can also happen in the GUI version, but you will have been
presented on screen 3 with a list of data types (FP SIGFP) and columns
from the input data file to match them and it is not so likely that you
will have chosen to match FP with IMEAN.)
>>
>> You can tell if this has happened in your case by looking for
>> something like (search for the string "Input labels:"):
>>
>> Input labels: ['IMEAN', 'SIGIMEAN', 'None', 'None', 'None', 'None',
'None', 'None', 'FreeR_flag']
>>
>> in your log file (where the None's don't matter, just the first 2
column names.  The columns in this list should be
>> amplitude-containing columns only.  If everything is ok.it should look
instead like:
>>
>> Input labels: ['F', 'SIGF', 'None', 'None', 'None', 'None', 'None',
'None', 'None']
>>
>> THE SOLUTION:
>>
>> In existing versions, the solution to this is to either (1) only have
amplitude arrays in your mtz file, or (2) use the GUI and select the
correct amplitude arrays, or (3) specify what amplitude columns to use
with a command like:
>>
>> input_labels="F SIGF".
>>
>> Note that you cannot use intensity data from an mtz file with the
existing versions of autobuild, only amplitude data. In the corrected
version (1.3b/rc7, not yet out) , you will be able to use an mtz file
with intensity data as long as there is no amplitude data in the same
mtz file.
>>
>> Note also that all this applies equally to the refinement_file and
hires_file data.  Also note that if the file type is not mtz, then this
bug does not apply (because the file is always imported and converted
from intensities to amplitudes automatically).
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Thomas C. Terwilliger
>> Mail Stop M888
>> Los Alamos National Laboratory
>> Los Alamos, NM 87545
>>
>> Tel:  505-667-0072                 email: terwilliger at LANL.gov
>> Fax: 505-665-3024                 SOLVE web site: http://solve.lanl.gov
PHENIX web site: http:www.phenix-online.org
>> ISFI Integrated Center for Structure and Function Innovation web site:
http://techcenter.mbi.ucla.edu
>> TB Structural Genomics Consortium web site:
>> http://www.doe-mbi.ucla.edu/TB
>>
>>
>> _______________________________________________
>> phenixbb mailing list
>> phenixbb at phenix-online.org
>> http://www.phenix-online.org/mailman/listinfo/phenixbb
>>
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://www.phenix-online.org/mailman/listinfo/phenixbb
>