phenix.phaser params.eff labin problem
Dear all, I am now writing a script for phenix.phaser. In the input file params.eff there is a keyword labin which corresponds to the labels inside input mtz files. The important columns in my ccp4 mtz file are FP, SIGFP. I edit the line of labin as: labin F=FP SIGF=SIGFP when running the script, it returns me: Syntax error: expected "=", found "F" (file "params.eff", line 9) So, I changed it like: labin=FP SIGFP when running, it gave me the message: Sorry: Couldn't find array FP SIGP in file I also tried: labin="FP" or labin = "FP SIGFP" or labin="labin F=FP" and so on. With nothing it can find the right columns. My mtz file should be not problem, I checked with mtzdmp, the columns FP and SIGFP were there and with phenix GUI this file works well. Would someone know exactly how to edit this line in params.eff? Many thanks! GP
On Tue, Jul 19, 2011 at 3:13 AM, G P
labin="FP,SIGFP" The syntax used for these parameter files is much different from the CCP4 keyword input - this was necessary for incorporating Phaser into the Phenix GUI. Among other things, we specify data labels in groups, and internally "FP,SIGFP" is a single data structure containing both the amplitudes and sigmas. FYI, unlike the CCP4 version, with phenix.phaser you can use any type of data as input, so an anomalous Scalepack file will work too. (Of course you still need to know what labels to use - someday I intend to make the setup more automatic on the command line.) -Nat
Thanks a lot Nat!
I tried exactly what you suggested. Still it returned me could not find the
labels. I also tried labin = "FP,SIGFP" and labin="FP, SIGFP" and
other possiblities with or withour space in between. They all returned me
could not find the label FP, SIGFP.
Should I give the keyword in capital letters. Or may be I should try other
format input file? I do not care about anomalous signal in this data,
because I will just use MR_AUTO model in phenix.phaser.
Any other ideas would be greatly appreciated. Thanks again for your time.
Best regards,
GP
On Tue, Jul 19, 2011 at 5:47 PM, Nathaniel Echols
Hi Nat,
Now I can make it. I just type labin="FP,SIGFP,DANO,SIGDANO". Then there is
no error to find the labels. I think because I should give all columns
together.
But there is another problem. When I run the script, it shows me:
...
File "/usr/local/phenix-1.7.1-743/phaser/phaser/phenix_interface/driver.py",
line 694, in __init__
assert os.path.isfile(params.phaser.solution)
AssertionError
File "/usr/local/phenix-1.7.1-743/phaser/phaser/phenix_interface/driver.py",
line 694, in __init__
assert os.path.isfile(params.phaser.solution)
AssertionError
Just stop at this step. What would be the reason and how to fix it? Any idea
would be greatly appreciated.
Best regards,
GP
On Tue, Jul 19, 2011 at 5:47 PM, Nathaniel Echols
On Wed, Jul 20, 2011 at 3:46 AM, G P
Right, sorry, that's another thing that CCTBX does automatically. This command will show you what array labels Phenix recognizes: iotbx.reflection_file_reader data.mtz
Look for a parameter called "solution" - this needs to be left as None. (I don't think there's any way to use this from the command line right now - it's very specific to the GUI.) -Nat
Hi Nat,
Thank you so much! That's cool. I just wrote a None to that solution and it
works!! Now I have some new problems:
At beginning, I only have phenix installed. I do not have CCP4 installed.
Then I run "phenix.phaser params.eff" to call phenix run phaser from script
params.eff. It works well.
Then, I installed CCP4 with its own phaser inside. Then I run again
"phenix.phaser params.eff". The computer start to import files (should be
some libraries and supporting files, not input data) from CCP4 installation
directory. It seems so strange for me.
And because some of the .so files in CCP4 directory are for 32bit,
phenix.phaser gave me error message: wrong ELF class: ELFCLASS32 and refused
to keep running.
Is there anyway to set some kind of PATH variables to let phenix.phaser
params.eff run like without CCP4 installed. (which I did at beginning).
Sorry for so many questions. Don't worry to reply if you do not have time.
I fully understand and thanks a lot for your reply.
regards,
GP
On Wed, Jul 20, 2011 at 8:27 PM, Nathaniel Echols
participants (3)
-
G P -
G Y
-
Nathaniel Echols