[phenixbb] on Phenix's "Generate "Table 1" for journal"
lodowski at gmail.com
Sat Mar 21 09:32:59 PDT 2015
Hi Smith, I thought that I might quickly answer your questions, But I think
Pavel is correct that some crystallographic reading will be instructive and
assist you in answering these questions. Is there someone in you laboratory
or at your institution that is more experienced that could assist you?
I'd suggest that inaddition to the Rupp text, Principles of Protein X-Ray
Crystallography. By J. Drenth is also quite good.
The short answer is that much or all of these data can be gathered from the
respective log files that are generated during processing/refinement or
from the header of the PDB file.
My responses follow your questions.
--First, for the "Total reflections" and "Multiplicity", they are empty (no
The total reflections for the processing section for table 1, you can get
these numbers from the HKL2000 scaling log, or if you used aimless, they
appear in a chart after the initial creation of the mtz file. I'm guessing
that you didn't include the logs during the running of the generate table 1
program (as Pavel has suggested), but truthfully you should be able to
locate these numbers from the log files of the programs, if they didn't
propagate to the final table.
--Secondly, it writes, "Statistics for the highest-resolution shell are
shown in parentheses". Suppose my crystal resolution is 2.5-45.6, then the the
highest-resolution value should be 2.5, rather than 45.6, right?
No, your highest shell would be something like all reflections between
2.6 and 2.5 A
-- In addition, what is the diffrence between highest-resolution shell in
the Table 1 and highest-resolution bin in the "Results" of Phenix refine?
I don't believe that the shells are necessarily the same ranges, and in
some cases you might have slightly different resolutions for processing and
refinement. depending on your program that you use, you can bin the data
into any number of bins -- for high high resolution data phenix has many
more "bins" in the results section, but when you processed in
--What is R-merge, R-meas,CC1/2 and CC*? I can get these data from MOSFLM
output mtz file. But is my images were processe by HKL, which I have no
experience, then how can I get the R-merge, R-meas,CC1/2 and CC*?
If you processed data, these numbers are diagnostic of the quality of the
dataset (R-merge/r-meas/R-pim) and (CC* and CC1/2, or the more traditional
I/sigma=2) indicate where data should be cut off. what basis did you you
use for where to cut off your data for refinement? When you run the
scalepack portion of HKL2000, it makes a log file with appropriate tables
at the end, detailing all of these numbers. newer versions of HKL2000 do
output CC1/2 along with I/sigma values to assist you in where the useful
range of the data ends, again found in the log file.
Hope this helps,
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the phenixbb