- phenixbb - phenix-online.org

Rfree % calculation
by Tarek DawoD 22 Jan '15

22 Jan '15

Dear ALL when you start you refinment in phenix with ***.sca file, fraction of reflection in test set will be created. under options (bottom) in phenix refine, you could see that fraction of reflection in test set is 0.1 which means 10 %by looking to the PDB header for completely different structures (below) you will find different percentage but always FREE R VALUE TEST SET COUNT is around 3800. so it always take the same 3800 count regardless the number of reflections. when you look in generated mtz file after the refinment, you see under R-free-flags(+) and R-free-flags(-) either 0 or 1. what does 0 or 1 mean? thank you in advance for explaination. TKDawoD ----------------------------------------------------------------------------------------------------REMARK 3 NUMBER OF REFLECTIONS : 47312 REMARK 3 NUMBER OF REFLECTIONS (NON-ANOMALOUS) : 24389 REMARK 3 REMARK 3 FIT TO DATA USED IN REFINEMENT.REMARK 3 R VALUE (WORKING + TEST SET) : 0.1673REMARK 3 R VALUE (WORKING SET) : 0.1636REMARK 3 FREE R VALUE : 0.2084REMARK 3 FREE R VALUE TEST SET SIZE (%) : 8.21 REMARK 3 FREE R VALUE TEST SET COUNT : 3882------------------------------------------------------------------------------------------------------REMARK 3 NUMBER OF REFLECTIONS : 107005 REMARK 3 NUMBER OF REFLECTIONS (NON-ANOMALOUS) : 56251 REMARK 3 REMARK 3 FIT TO DATA USED IN REFINEMENT.REMARK 3 R VALUE (WORKING + TEST SET) : 0.1457REMARK 3 R VALUE (WORKING SET) : 0.15.34REMARK 3 FREE R VALUE : 0.1997REMARK 3 FREE R VALUE TEST SET SIZE (%) : 3.51 REMARK 3 FREE R VALUE TEST SET COUNT : 3809-----------------------------------------------------------------------------------------------------NUMBER OF REFLECTIONS : 48150 REMARK 3 NUMBER OF REFLECTIONS (NON-ANOMALOUS) : 25296 REMARK 3 REMARK 3 FIT TO DATA USED IN REFINEMENT.REMARK 3 R VALUE (WORKING + TEST SET) : 0.1507REMARK 3 R VALUE (WORKING SET) : 0.1470REMARK 3 FREE R VALUE : 0.1918REMARK 3 FREE R VALUE TEST SET SIZE (%) : 7.98 REMARK 3 FREE R VALUE TEST SET COUNT : 3841

3 3

Problems installing Rosetta
by 신인철 20 Jan '15

20 Jan '15

Hi, I am trying to install rosetta on Fedora 21. I downloaded rosetta_bin_linux_2015.02.57538_bundle.tgz and copied the unpacked whole directory under /usr/local/ and rename "rosetta_bin_linux_2015.02.57538" as "rosetta" simply. And I edited .cshrc file like: setenv PHENIX_ROSETTA_PATH "/usr/local/rosetta" setenv ROSETTA3_DB "usr/local/rosetta/main/database" But encountered the following messages: [root@localhost ~]# rosetta.build_phenix_interface nproc=8 False scons: Reading SConscript files ... Running versioning script ... Done. (0.0 seconds) Could not find platform independent libraries <prefix> Could not find platform dependent libraries <exec_prefix> Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>] ImportError: No module named site CalledProcessError: Command '['./update_options.sh']' returned non-zero exit status 1: File "/usr/local/rosetta/main/source/SConstruct", line 172: main() File "/usr/local/rosetta/main/source/SConstruct", line 163: SConscript("SConscript", "build") File "/usr/local/rosetta/main/source/external/scons-local/scons-local-2.0.1/SCons/Script/SConscript.py", line 614: return method(*args, **kw) File "/usr/local/rosetta/main/source/external/scons-local/scons-local-2.0.1/SCons/Script/SConscript.py", line 551: return _SConscript(self.fs, *files, **subst_kw) File "/usr/local/rosetta/main/source/external/scons-local/scons-local-2.0.1/SCons/Script/SConscript.py", line 260: exec _file_ in call_stack[-1].globals File "/usr/local/rosetta/main/source/SConscript", line 28: subprocess.check_call(["./update_options.sh"]) File "/usr/local/phenix-1.9-1692/build/intel-linux-2.6-x86_64/base/lib/python2.7/subprocess.py", line 540: raise CalledProcessError(retcode, cmd) What am I going to do? Please, help!! ------------------------------------------------- Inchul Shin School of Agricultural Biotechnology Seoul National University Seoul, KOREA 151-921 Tel: +82-2-880-4633 Fax: +82-2-873-4950 Mobile: +82-10-6215-2896 Email: ichul96(a)snu.ac.kr<mailto:[email protected]>

2 1

Font woes in Debian with latest nightly builds
by Miguel Ortiz Lombardía 20 Jan '15

20 Jan '15

Hi, I recently updated from dev-1805 to dev-1924 (in both cases I used CentOS5 installers for a Debian testing -jessie- machine; dev-1922 gave the same problem) During installation I observed these messages: <...> updating fonts/local.conf file running mkfontdir Couldn't determine weight for Vera.pfa Couldn't determine weight for VeraMono.pfa rebuilding font cache Fontconfig error: Cannot load default config file /usr/local/phenix-dev-1922/base/bin/fc-cache: "/usr/local/phenix-dev-1922/base/share/fonts": skipping, no such directory /usr/local/phenix-dev-1922/base/bin/fc-cache: succeeded <...> I found a previous post to this list with a similar report (january 29, 2013) but there it was said that these messages could be ignored... However, when I run phenix I see these in the terminal: $ phenix Linux /usr/local/phenix-dev-1922/build/bin/phenix: 109: [: Linux: unexpected operator Fontconfig error: Cannot load default config file Phenix data directory: /home/ortiz-lombardia/Phenix (python2.7:11563): Pango-WARNING **: No builtin or dynamically loaded modules were found. Pango will not work correctly. This probably means there was an error in the creation of: '/home/builder/slave/phenix-nightly-intel-linux-2_6-x86_64-centos5/base/etc/pango/pango.modules' You should create this file by running pango-querymodules. (python2.7:11563): Pango-WARNING **: pango_shape called with bad font, expect ugly output (python2.7:11563): Pango-WARNING **: pango_font_get_glyph_extents called with null font argument, expect ugly output (python2.7:11563): Pango-WARNING **: pango_font_get_metrics called with null font argument, expect ugly output (python2.7:11563): GdkPixbuf-WARNING **: Cannot open pixbuf loader module file '/home/builder/slave/phenix-nightly-intel-linux-2_6-x86_64-centos5/base/etc/gtk-2.0/gdk-pixbuf.loaders': No such file or directory (python2.7:11563): GdkPixbuf-WARNING **: Cannot open pixbuf loader module file '/home/builder/slave/phenix-nightly-intel-linux-2_6-x86_64-centos5/base/etc/gtk-2.0/gdk-pixbuf.loaders': No such file or directory (python2.7:11563): Gtk-WARNING **: Error loading theme icon for stock: Couldn't recognize the image file format for file '/usr/share/icons/gnome/48x48/status/gtk-dialog-error.png' (python2.7:11563): Pango-WARNING **: _pango_cairo_font_install called with bad font, expect ugly output ^C Indeed, it is ugly output... No readable text just little empty boxes. I can send a snapshot if required. If I run: dpkg -l \*pango\* un gir1.0-pango-1 <none> <none> (no description available) ii gir1.2-coglpan 1.18.2-3 amd64 GObject introspection data for th ii gir1.2-pango-1 1.36.8-3 amd64 Layout and rendering of internati ii libcogl-pango2 1.18.2-3 amd64 Object oriented GL/GLES Abstracti ii libpango-1.0-0 1.36.8-3 amd64 Layout and rendering of internati ii libpango-perl 1.226-2 amd64 Perl module to layout and render ii libpango1.0-0: 1.36.8-3 amd64 Layout and rendering of internati ii libpango1.0-de 1.36.8-3 amd64 Development files for the Pango un libpango1.0-do <none> <none> (no description available) ii libpangocairo- 1.36.8-3 amd64 Layout and rendering of internati ii libpangoft2-1. 1.36.8-3 amd64 Layout and rendering of internati ii libpangomm-1.4 2.34.0-1.1 amd64 C++ Wrapper for pango (shared lib ii libpangox-1.0- 0.0.2-5 amd64 pango library X backend ii libpangox-1.0- 0.0.2-5 amd64 pango library X backend - develop ii libpangoxft-1. 1.36.8-3 amd64 Layout and rendering of internati un pango-graphite <none> <none> (no description available) un pango1.0-multi <none> <none> (no description available) Also, dpkg -l \*fontconfig\* produces: ii fontconfig 2.11.0-6.3 amd64 generic font configuration librar ii fontconfig-con 2.11.0-6.3 all generic font configuration librar un libfontconfig <none> <none> (no description available) un libfontconfig- <none> <none> (no description available) ii libfontconfig1 2.11.0-6.3 amd64 generic font configuration librar ii libfontconfig1 2.11.0-6.3 i386 generic font configuration librar ii libfontconfig1 2.11.0-6.3 amd64 generic font configuration librar I would appreciate any clues to solve this issue. Cheers, Miguel

2 1

Bug in phenix.carbo_load in latest nightlies
by Roger Dodd 20 Jan '15

20 Jan '15

Hi all, I've been using the carbo_load tool to add in GlcNAc residues to Asn side chains and have noticed that in recent nightlies there is a bug that throws up the following error message (this occurs even if carbo_load is just run in "mode 1"): Traceback (most recent call last): File "/Applications/phenix-dev-1924/build/../modules/elbow/elbow/command_line/carbo_load.py", line 705, in <module> run(args) File "/Applications/phenix-dev-1924/build/../modules/elbow/elbow/command_line/carbo_load.py", line 417, in run from elbow.saccharide import carbohydrate_classes File "/Applications/phenix-dev-1924/modules/elbow/elbow/saccharide/carbohydrate_classes.py", line 28, in <module> from phenix.ligands.DensityBlobsClassConnectivity import DensityBlobsClassConnectivity ImportError: No module named DensityBlobsClassConnectivity I get this with dev-1894 and dev-1924 but not the last stable 1.9-1692. Cheers, Roger -- Roger B. Dodd, PhD Cambridge Institute for Medical Research University of Cambridge School of Clinical Medicine Wellcome Trust/MRC Building Cambridge Biomedical Campus Hills Road Cambridge CB2 0XY

2 1

Molecular replacement with fixed solution
by Muhammed bashir Khan 19 Jan '15

19 Jan '15

Hi There; I want to solve the structure with molecular replacement using phaser in phenix GUI. When I get solution from the one, how I can use as it as already solution using phenix in the next run. I use the "ensamble is fixed partial solution" but seems its not working. Any suggestion will be highly appreciated. Thanks Regards -- Muhammad Bashir Khan ************************************************** Department for Structural and Computational Biology Max F. Perutz Laboratories University of Vienna Campus Vienna Biocenter 5 A-1030 Vienna Austria Austria Phone: +43(1)427752224 Fax: +43(1)42779522

2 2

sudden increase of clashscore
by Smith Lee 19 Jan '15

19 Jan '15

Dear All, In a Phenix refinement, with refinement settings as default in the graphics interface, at 2' rsrl stage the clashscore suddenly increases to more than 50, and then gradually decreases to lower than 20. Will you please tell me why this can occur? And what its effect on the refinement? Smith

1 0

Re: [phenixbb] questions related to Phenix refinement
by Kay Diederichs 18 Jan '15

18 Jan '15

Dear Pavel, Am 18.01.15 um 07:23 schrieb phenixbb-request(a)phenix-online.org: > Date: Sat, 17 Jan 2015 20:36:30 -0800 > From: Pavel Afonine <pafonine(a)lbl.gov> > To: Smith Lee <smith_lee123(a)yahoo.com>, "phenixbb(a)phenix-online.org" > <phenixbb(a)phenix-online.org> > Subject: Re: [phenixbb] questions related to Phenix refinement > Message-ID: <54BB384E.30602(a)lbl.gov> > Content-Type: text/plain; charset="windows-1252"; Format="flowed" > > Hello Smith, > >> > In the X-ray statistics by resolution bin of the Phenix.refine result, >> > there is a column "%complete". For my refinement data, I find the >> > better the resolution (from lower resolution to the higher >> > resolution), the lower the completeness (for example for 40-6 A, >> > %complete is 98, for 3.1-3.0 A, %complete is 60%, for 2.2-2.1 A, >> > %complete is 6%). >> > >> > Will you please tell me what does this "%complete" mean? why it >> > decreases in the better diffraction bin? > Completeness is how many reflections you have compared to theoretically > possible. So the higher completeness the better. Ideally (and it's not > that uncommon these days) you should have 100% complete data set in > d_min-inf resolution. Anything below say 80 in any resolution bin is > bad, and numbers you quote 6-60% mean something is wrong withe the dataset. > Given your standing in the community, the last sentence will lead many unexperienced people to believe that they should cut their data at the resolution where the completeness falls below "say 80"%. But that would be wrong. There is no reason to consider a completeness as "too low in a high-resolution shell" as long as the data in that shell are good. Particularly in refinement any reflection helps to improve the model, and to reduce overfitting. Of course, more complete is better! Nowadays there should be no reason _not_ to get >95% (>99% in high-symmetry spacegroups) completeness in the low resolution shells, unless people stick to the "collect minimum rotation range" paradigm, and get the starting angle (or the point group!) wrong. best, Kay

3 3

FW: question regarding Table1
by SD Y 18 Jan '15

18 Jan '15

From: ccp4_sc(a)hotmail.com To: phenixbb-request(a)phenix-online.org Subject: question regarding Table1 Date: Fri, 16 Jan 2015 20:30:47 -0600 Hi Everyone, I used phenix refined pdb, mtz and original mtz in the latest version of phenix to generate Table 1 and I felt the following was not right. I am not sure of the reason. R-merge 1.193e-17 (1.036e-17) R-meas 1.688e-17 CC1/2 1 (1) CC* 1 (1) Let me know . I appreciate your comments. Regards,SDY

3 2

questions related to Phenix refinement
by Smith Lee 18 Jan '15

18 Jan '15

Dear All, In the X-ray statistics by resolution bin of the Phenix.refine result, there is a column "%complete". For my refinement data, I find the better the resolution (from lower resolution to the higher resolution), the lower the completeness (for example for 40-6 A, %complete is 98, for 3.1-3.0 A, %complete is 60%, for 2.2-2.1 A, %complete is 6%). Will you please tell me what does this "%complete" mean? why it decreases in the better diffraction bin? For the Ramachandran restrain in the Phenix.refine, there was a recommendation that for good resolution (I forget how much exactly, maybe 2.5 A), do not use Ramachandran restrain. But for a crystal data with resolution better than 1.5 A, in the X-ray statistics by resolution bin of the Phenix.refine result, there are more rows with resolution poorer than the defined 2.5 A (for example 40-6 A,6-3 A), although the crystal has a resolution better than 1.5 A (for example 1.4 A), can I use Ramachandran restrain in the Phenix.refine in this situation? If I use Ramachandran restrain in the Phenix.refine, I find the R-work and R-free level goes up, can we say the strategy of Ramachandran restrain in my refinement was not appropriate? I am looking forward to getting your reply. Smith

2 1

Xray / neutron joint refinement
by Anna Makal 16 Jan '15

16 Jan '15

Hi, I have just done a joint x-ray / neutron refinement in Phenix using 5PTI structure and data from PDB. The input contains H atoms, with positions as refined against neutron dataset. In the resulting model, all H atoms ended up in 'idealized X-ray' positions (C-H distance of 0.96A), just as they would if I performed refinement against xray data only with riding model for H hydrogen atoms. Is it possible to run joint X-ray neutron refinement in such a manner that final H atoms positions are based on neutron data? Regards, Anna Makal

3 2