Search results for query "look through"
- 527 messages
Re: [phenixbb] help with phenix.ligand_identification
by Edward A. Berry
Ligandfit itself has more helpful error message:
Sorry the string 'S2063' cannot be interpreted as a residue number?
(Duh!)
and does actually place the ligand in the requested blob with "search_center="
So I can run ligandfit in a foreach loop with each of the 180 ligands left by ligand_identification
=================
On 09/26/2017 03:32 PM, Edward A. Berry wrote:
> I'm having some problems using ligand_identification.
> I would like to restrict the search to a specific density peak, even if it is not the highest unmodeled peak or the highest Fo-Fc peak in the map.
> I tried using options:
> search_center="67.5 18.3 11.8"
> or
> ligand_near_res=S2063,
> Will the search be restricted to that region, or if a particuar ligand doesn't fit that blob,
> will it search through the rest of the map? If it is restricted, in what radius?
> Is this radius affected by the "search_dist" or "local_search" parameters?
> and local_search = True is default?
> Is there a threshold level for density level, below which building a ligand in a blob will not be attempted?
>
> I've tried with both search_center= and ligand_near_res=, and something gets
> built far away from that site. But there were errors, so I may have something wrong:
>
>
> phenix.ligand_identification mtz_in=sqr2803or13_031.mtz input_labels="2FOFCWT PH2FOFCWT" \
> model=sqr2803or13_031.pdb ligand_near_res=S2063 nproc=2
>
> After preparing the ligand library, then:
> Running LigandFit process 1...
>
> Number of atoms in ligand suc.pdb is 23
> Running job sequence 1, ligand 2, in /tb/sb/usr20c/berry/ref/sqrdep/sqr2803dep2/phenix2/Temp_1...
> Evaluating all ligands in ligand-lib now...and placing fittedligand ### in resolve_ligand_###.pdb
>
> Number of atoms in ligand 2pe.pdb is 28
> Running job sequence 0, ligand 1, in /tb/sb/usr20c/berry/ref/sqrdep/sqr2803dep2/phenix2/Temp_0...
> Process Process-2:
> Traceback (most recent call last):
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
> self.run()
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/multiprocessing/process.py", line 114, in run
> self._target(*self._args, **self._kwargs)
> File "/sw/lnx/phenix-1.12-2829/build/../modules/phenix/phenix/command_line/ligand_identification.py", line 1382, in RunLigandFit
> shutil.rmtree(ligandfit_dir)
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/shutil.py", line 247, in rmtree
> rmtree(fullname, ignore_errors, onerror)
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/shutil.py", line 256, in rmtree
> onerror(os.rmdir, path, sys.exc_info())
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/shutil.py", line 254, in rmtree
> os.rmdir(path)
> OSError: [Errno 39] Directory not empty: '/tb/sb/usr20c/berry/ref/sqrdep/sqr2803dep2/phenix2/Temp_1/LigandFit_run_1_/TEMP0'
> Process Process-1:
> Traceback (most recent call last):
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
> self.run()
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/multiprocessing/process.py", line 114, in run
> self._target(*self._args, **self._kwargs)
> File "/sw/lnx/phenix-1.12-2829/build/../modules/phenix/phenix/command_line/ligand_identification.py", line 1382, in RunLigandFit
> shutil.rmtree(ligandfit_dir)
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/shutil.py", line 247, in rmtree
> rmtree(fullname, ignore_errors, onerror)
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/shutil.py", line 256, in rmtree
> onerror(os.rmdir, path, sys.exc_info())
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/shutil.py", line 254, in rmtree
> os.rmdir(path)
> OSError: [Errno 39] Directory not empty: '/tb/sb/usr20c/berry/ref/sqrdep/sqr2803dep2/phenix2/Temp_0/LigandFit_run_1_/TEMP0'
>
> Evaluating LigandFit results ...
>
> The run continues, but it does not test any more ligands after those first two but goes on to evaluate the results.
> With nproc = 1, only 1 ligand gets tested. In all cases the first ligand in the library (2PE.pdb) is evaluated as the best.
> It is placed in density, but density that has already been built out with (and looks more like) a string of water molecules.
> And this is far from the selected residue or coordinates specified.
>
> The directory that raised the error when attempting to be deleted does eventually get removed: after the run there is no TEMP_N in the parent directory.
>
> Any suggestions would be welcome.
> Ed
>
> P.S.
> - running with .eff file:
>
>
> ['--show_defaults']
> ligand_identification {
> mtz_in = sqr2803or13_031.mtz
> mtz_type = *F diffmap
> model = sqr2803or13_031.pdb
> ncpu = 1
> n_indiv_tries_min = 30
> n_indiv_tries_max = 300
> n_group_search = 4
> search_dist = 10
> local_search = True
> search_center = "67.5 18.3 11.8"
> # ligand_near_res = S2063
> verbose = False
> debug = False
> use_ligandfit = True
> search_mode = *default LigandFit
> temp_dir = Auto
> dry_run = False
> # number_of_ligands = 1
> cc_min = 0.75
> open_in_coot = False
> non_bonded = True
> keep_all_files = False
> # cif_def_file_list =
> real_space_target_weight = 10
> # job_title = None
> ligandfit {
> }
> }
>
> gives:
>
> [['2pe.pdb', 'suc.pdb', . . . 'upl.pdb']]
>
> /tb/sb/usr20c/berry/ref/sqrdep/sqr2803dep2/phenix2
>
> *******************************************************************************
>
> Sorry, the protein model file None does not seem to exist?
>
> *******************************************************************************
>
>
> Running LigandFit process 0...
>
> Process Process-1:
> Traceback (most recent call last):
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
> self.run()
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/multiprocessing/process.py", line 114, in run
> self._target(*self._args, **self._kwargs)
> File "/sw/lnx/phenix-1.12-2829/build/../modules/phenix/phenix/command_line/ligand_identification.py", line 1122, in RunLigandFit
> shutil.copyfile(mtz_in,data_local)
> File "/sw/lnx/phenix-1.12-2829/base/lib/python2.7/shutil.py", line 82, in copyfile
> with open(src, 'rb') as fsrc:
> IOError: [Errno 2] No such file or directory: 'None'
>
> Evaluating LigandFit results ...
>
> Lig_seq Placed/total cc_all cc cc_adj score Code HBscore
>
> Cannot find overall_ligand_scores.log0. This could mean that none of that (sub)set of ligand fitted well.
>
>
>
> None of the ligand fit the difference desity well enough. Please try the following --
> 1) if you input a custom library, try to use the default library (no extra keywords needed), or
> 2) if you used the default library already, ususlly this means that the density is too small (> 6 atome or more is needed.)
> Exiting ......
>
>
>
> No good ligand found.
>
>
> _______________________________________________
> phenixbb mailing list
> phenixbb(a)phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave(a)phenix-online.org
>
8 years, 4 months
Re: [cctbxbb] some thoughts on cctbx and pip
by Luc Bourhis
Hi Graeme,
Yes, I know. But “black" is a program doing a very particular task (code formatting from the top of my head). Requiring to use a wrapper for python itself is another level. But ok, I think I am mellowing to the idea after all! Talking with people around me, and extrapolating, I would bet that, right now, a great majority of people interested by cctbx in pip have already used the cctbx, so they know about the Python wrapper, and they would not be too sanguine about that. My concern is for the future, when pip will be the first time some people use cctbx. Big fat warning notices on PyPI page and a better error message when cctbx fails because LIBTBX_BUILD is not set would be needed but that could be all right.
If we do a pip installer, we should aim at a minimal install: cctbx, iotbx and their dependencies, and that’s it.
Best wishes,
Luc
> On 23 Aug 2019, at 07:17, Graeme.Winter(a)Diamond.ac.uk <Graeme.Winter(a)diamond.ac.uk> wrote:
>
> Without discussing the merits of this or whether we _choose_ to make the move to supporting PIP, I am certain it would be _possible_ - many other packages make dispatcher scripts when you pip install them e.g.
>
> Silver-Surfer rescale_f2 :) $ which black; cat $(which black)
> /Library/Frameworks/Python.framework/Versions/3.6/bin/black
> #!/Library/Frameworks/Python.framework/Versions/3.6/bin/python3.6
>
> # -*- coding: utf-8 -*-
> import re
> import sys
>
> from black import main
>
> if __name__ == '__main__':
> sys.argv[0] = re.sub(r'(-script\.pyw?|\.exe)?$', '', sys.argv[0])
> sys.exit(main())
>
> So we _could_ work around the absence of LIBTBX_BUILD etc. in the system. Whether or not we elect to do the work is a different question, and it seems clear that here are very mixed opinions on this.
>
> Best wishes Graeme
>
>
> On 23 Aug 2019, at 01:21, Luc Bourhis <luc_j_bourhis(a)mac.com<mailto:[email protected]>> wrote:
>
> Hi,
>
> Even if we managed to ship our the boost dynamic libraries with pip, it would still not be pip-like, as we would still need our python wrappers to set LIBTBX_BUILD and LD_LIBRARY_PATH. Normal pip packages work with the standard python exe. LD_LIBRARY_PATH, we could get around that by changing the way we compile, using -Wl,-R, which is the runtime equivalent of build time -L. That’s a significant change that would need to be tested. But there is no way around setting LIBTBX_BUILD right now. Leaving that to the user is horrible. Perhaps there is a way to hack libtbx/env_config.py so that we can hardwire LIBTBX_BUILD in there when pip installs?
>
> Best wishes,
>
> Luc
>
>
> On 16 Aug 2019, at 22:47, Luc Bourhis <luc_j_bourhis(a)mac.com<mailto:[email protected]>> wrote:
>
> Hi,
>
> I did look into that many years ago, and even toyed with building a pip installer. What stopped me is the exact conclusion you reached too: the user would not have the pip experience he expects. You are right that it is a lot of effort but is it worth it? Considering that remark, I don’t think so. Now, Conda was created specifically to go beyond pip pure-python-only support. Since cctbx has garnered support for Conda, the best avenue imho is to go the extra length to have a package on Anaconda.org<http://anaconda.org/>, and then to advertise it hard to every potential user out there.
>
> Best wishes,
>
> Luc
>
>
> On 16 Aug 2019, at 21:45, Aaron Brewster <asbrewster(a)lbl.gov<mailto:[email protected]>> wrote:
>
> Hi, to avoid clouding Dorothee's documentation email thread, which I think is a highly useful enterprise, here's some thoughts about putting cctbx into pip. Pip doesn't install non-python dependencies well. I don't think boost is available as a package on pip (at least the package version we use). wxPython4 isn't portable through pip (https://wiki.wxpython.org/How%20to%20install%20wxPython#Installing_wxPython…). MPI libraries are system dependent. If cctbx were a pure python package, pip would be fine, but cctbx is not.
>
> All that said, we could build a manylinux1 version of cctbx and upload it to PyPi (I'm just learning about this). For a pip package to be portable (which is a requirement for cctbx), it needs to conform to PEP513, the manylinux1 standard (https://www.python.org/dev/peps/pep-0513/). For example, numpy is built according to this standard (see https://pypi.org/project/numpy/#files, where you'll see the manylinux1 wheel). Note, the manylinux1 standard is built with Centos 5.11 which we no longer support.
>
> There is also a manylinux2010 standard, which is based on Centos 6 (https://www.python.org/dev/peps/pep-0571/). This is likely a more attainable target (note though by default C++11 is not supported on Centos 6).
>
> If we built a manylinuxX version of cctbx and uploaded it to PyPi, the user would need all the non-python dependencies. There's no way to specify these in pip. For example, cctbx requires boost 1.63 or better. The user will need to have it in a place their python can find it, or we could package it ourselves and supply it, similar to how the pip h5py package now comes with an hd5f library, or how the pip numpy package includes an openblas library. We'd have to do the same for any packages we depend on that aren't on pip using the manylinux standards, such as wxPython4.
>
> Further, we need to think about how dials and other cctbx-based packages interact. If pip install cctbx is set up, how does pip install dials work, such that any dials shared libraries can find the cctbx libraries? Can shared libraries from one pip package link against libraries in another pip package? Would each package need to supply its own boost? Possibly this is well understood in the pip field, but not by me :)
>
> Finally, there's the option of providing a source pip package. This would require the full compiler toolchain for any given platform (macOS, linux, windows). These are likely available for developers, but not for general users.
>
> Anyway, these are some of the obstacles. Not saying it isn't possible, it's just a lot of effort.
>
> Thanks,
> -Aaron
>
> _______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org<mailto:[email protected]>
> http://phenix-online.org/mailman/listinfo/cctbxbb
>
> _______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org<mailto:[email protected]>
> http://phenix-online.org/mailman/listinfo/cctbxbb
>
> _______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org<mailto:[email protected]>
> http://phenix-online.org/mailman/listinfo/cctbxbb
>
>
> --
> This e-mail and any attachments may contain confidential, copyright and or privileged material, and are for the use of the intended addressee only. If you are not the intended addressee or an authorised recipient of the addressee please notify us of receipt by returning the e-mail and do not use, copy, retain, distribute or disclose the information in or attached to the e-mail.
> Any opinions expressed within this e-mail are those of the individual and not necessarily of Diamond Light Source Ltd.
> Diamond Light Source Ltd. cannot guarantee that this e-mail or any attachments are free from viruses and we cannot accept liability for any damage which you may sustain as a result of software viruses which may be transmitted in or with the message.
> Diamond Light Source Limited (company no. 4375679). Registered in England and Wales with its registered office at Diamond House, Harwell Science and Innovation Campus, Didcot, Oxfordshire, OX11 0DE, United Kingdom
>
> _______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org
> http://phenix-online.org/mailman/listinfo/cctbxbb
6 years, 5 months
Re: [phenixbb] Geometry Restraints - Anisotropic truncation
by Frank von Delft
Is the explanation not simpler? The volumes of reciprocal space that
were left out did not in fact contain signal, and it's by removing those
noise non-reflections that the actual R is revealed. As James Holton
posted a while ago, Rfactors calculated for noise give randomly large
values.
So it seems less less misleading to refer to it as "anisotropy-TRUNCATED".
phx.
On 03/05/2012 07:40, Pavel Afonine wrote:
> Hi Kendall,
>
> I just did this quick test: calculated R-factors using original and
> anisotropy-corrected Mike Sawaya's data (*)
>
> Original:
> r_work : 0.3026
> r_free : 0.3591
> number of reflections: 26944
>
> Truncated:
> r_work : 0.2640
> r_free : 0.3178
> number of reflections: 18176
>
> The difference in R-factors is not too surprising given how many
> reflections was removed (about 33%).
>
> Pavel
>
> (*) Note, the data available in PDB is anisotropy corrected. The
> original data set was kindly provided to me by the author.
>
>
> On 5/2/12 5:25 AM, Kendall Nettles wrote:
>> I didnt think the structure was publishable with Rfree of 33%
>> because I was expecting the reviewers to complain.
>>
>> We have tested a number of data sets on the UCLA server and it
>> usually doesn't make much difference. I wouldn't expect truncation
>> alone to change Rfree by 5%, and it usually doesn't. The two times I
>> have seen dramatic impacts on the maps ( and Rfree ), the highly
>> anisotrophic sets showed strong waves of difference density as well,
>> which was fixed by throwing out the noise. We have moved to using
>> loose data cutoffs for most structures, but I do think anisotropic
>> truncation can be helpful in rare cases.
>>
>> Kendall
>>
>> On May 1, 2012, at 3:07 PM, "Dale
>> Tronrud"<det102(a)uoxray.uoregon.edu> wrote:
>>
>>> While philosophically I see no difference between a spherical
>>> resolution
>>> cutoff and an elliptical one, a drop in the free R can't be the
>>> justification
>>> for the switch. A model cannot be made more "publishable" simply by
>>> discarding
>>> data.
>>>
>>> We have a whole bunch of empirical guides for judging the
>>> quality of this
>>> and that in our field. We determine the resolution limit of a data
>>> set (and
>>> imposing a "limit" is another empirical choice made) based on Rmrg,
>>> or Rmes,
>>> or Rpim getting too big or I/sigI getting too small and there is no
>>> agreement
>>> on how "too big/small" is too "too big/small".
>>>
>>> We then have other empirical guides for judging the quality of
>>> the models
>>> we produce (e.g. Rwork, Rfree, rmsds of various sorts). Most people
>>> seem to
>>> recognize that the these criteria need to be applied differently for
>>> different
>>> resolutions. A lower resolution model is allowed a higher Rfree, for
>>> example.
>>>
>>> Isn't is also true that a model refined to data with a cutoff of
>>> I/sigI of
>>> 1 would be expected to have a free R higher than a model refined to
>>> data with
>>> a cutoff of 2? Surely we cannot say that the decrease in free R
>>> that results
>>> from changing the cutoff criteria from 1 to 2 reflects an improved
>>> model. It
>>> is the same model after all.
>>>
>>> Sometimes this shifting application of empirical criteria
>>> enhances the
>>> adoption of new technology. Certainly the TLS parametrization of
>>> atomic
>>> motion has been widely accepted because it results in lower working
>>> and free
>>> Rs. I've seen it knock 3 to 5 percent off, and while that certainly
>>> means
>>> that the model fits the data better, I'm not sure that the quality
>>> of the
>>> hydrogen bond distances, van der Waals distances, or maps are any
>>> better.
>>> The latter details are what I really look for in a model.
>>>
>>> On the other hand, there has been good evidence through the
>>> years that
>>> there is useful information in the data beyond an I/sigI of 2 or an
>>> Rmeas> 100% but getting people to use this data has been a hard
>>> slog. The
>>> reason for this reluctance is that the R values of the resulting models
>>> are higher. Of course they are higher! That does not mean the models
>>> are of poorer quality, only that data with lower signal/noise has been
>>> used that was discarded in the models you used to develop your "gut
>>> feeling"
>>> for the meaning of R.
>>>
>>> When you change your criteria for selecting data you have to
>>> discard
>>> your old notions about the acceptable values of empirical quality
>>> measures.
>>> You either have to normalize your measure, as Phil Jeffrey
>>> recommends, by
>>> ensuring that you calculate your R's with the same reflections, or by
>>> making objective measures of map quality.
>>>
>>> Dale Tronrud
>>>
>>> P.S. It is entirely possible that refining a model to a very optimistic
>>> resolution cutoff and calculating the map to a lower resolution
>>> might be
>>> better than throwing out the data altogether.
>>>
>>> On 5/1/2012 10:34 AM, Kendall Nettles wrote:
>>>> I have seen dramatic improvements in maps and behavior during
>>>> refinement following use of the UCLA anisotropy server in two
>>>> different cases. For one of them the Rfree went from 33% to 28%. I
>>>> don't think it would have been publishable otherwise.
>>>> Kendall
>>>>
>>>> On May 1, 2012, at 11:10 AM, Bryan Lepore wrote:
>>>>
>>>>> On Mon, Apr 30, 2012 at 4:22 AM, Phil
>>>>> Evans<pre(a)mrc-lmb.cam.ac.uk> wrote:
>>>>>> Are anisotropic cutoff desirable?
>>>>> is there a peer-reviewed publication - perhaps from Acta
>>>>> Crystallographica - which describes precisely why scaling or
>>>>> refinement programs are inadequate to ameliorate the problem of
>>>>> anisotropy, and argues why the method applied in Strong, et. al. 2006
>>>>> satisfies this need?
>>>>>
>>>>> -Bryan
>
> _______________________________________________
> phenixbb mailing list
> phenixbb(a)phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
13 years, 9 months
Re: [phenixbb] Geometry Restraints - Anisotropic truncation
by Pavel Afonine
Hi Kendall,
removing same amount of data randomly gives Rwork/Rfree ~ 30/35%.
Pavel
On 5/3/12 4:13 AM, Kendall Nettles wrote:
> Hi Pavel,
> What happens if you throw out that many reflections that have signal? Can you take out a random set of the same size?
> Best,
> Kendall
>
> On May 3, 2012, at 2:41 AM, "Pavel Afonine"<pafonine(a)lbl.gov> wrote:
>
>> Hi Kendall,
>>
>> I just did this quick test: calculated R-factors using original and
>> anisotropy-corrected Mike Sawaya's data (*)
>>
>> Original:
>> r_work : 0.3026
>> r_free : 0.3591
>> number of reflections: 26944
>>
>> Truncated:
>> r_work : 0.2640
>> r_free : 0.3178
>> number of reflections: 18176
>>
>> The difference in R-factors is not too surprising given how many
>> reflections was removed (about 33%).
>>
>> Pavel
>>
>> (*) Note, the data available in PDB is anisotropy corrected. The
>> original data set was kindly provided to me by the author.
>>
>>
>> On 5/2/12 5:25 AM, Kendall Nettles wrote:
>>> I didnt think the structure was publishable with Rfree of 33% because I was expecting the reviewers to complain.
>>>
>>> We have tested a number of data sets on the UCLA server and it usually doesn't make much difference. I wouldn't expect truncation alone to change Rfree by 5%, and it usually doesn't. The two times I have seen dramatic impacts on the maps ( and Rfree ), the highly anisotrophic sets showed strong waves of difference density as well, which was fixed by throwing out the noise. We have moved to using loose data cutoffs for most structures, but I do think anisotropic truncation can be helpful in rare cases.
>>>
>>> Kendall
>>>
>>> On May 1, 2012, at 3:07 PM, "Dale Tronrud"<det102(a)uoxray.uoregon.edu> wrote:
>>>
>>>> While philosophically I see no difference between a spherical resolution
>>>> cutoff and an elliptical one, a drop in the free R can't be the justification
>>>> for the switch. A model cannot be made more "publishable" simply by discarding
>>>> data.
>>>>
>>>> We have a whole bunch of empirical guides for judging the quality of this
>>>> and that in our field. We determine the resolution limit of a data set (and
>>>> imposing a "limit" is another empirical choice made) based on Rmrg, or Rmes,
>>>> or Rpim getting too big or I/sigI getting too small and there is no agreement
>>>> on how "too big/small" is too "too big/small".
>>>>
>>>> We then have other empirical guides for judging the quality of the models
>>>> we produce (e.g. Rwork, Rfree, rmsds of various sorts). Most people seem to
>>>> recognize that the these criteria need to be applied differently for different
>>>> resolutions. A lower resolution model is allowed a higher Rfree, for example.
>>>>
>>>> Isn't is also true that a model refined to data with a cutoff of I/sigI of
>>>> 1 would be expected to have a free R higher than a model refined to data with
>>>> a cutoff of 2? Surely we cannot say that the decrease in free R that results
>>>> from changing the cutoff criteria from 1 to 2 reflects an improved model. It
>>>> is the same model after all.
>>>>
>>>> Sometimes this shifting application of empirical criteria enhances the
>>>> adoption of new technology. Certainly the TLS parametrization of atomic
>>>> motion has been widely accepted because it results in lower working and free
>>>> Rs. I've seen it knock 3 to 5 percent off, and while that certainly means
>>>> that the model fits the data better, I'm not sure that the quality of the
>>>> hydrogen bond distances, van der Waals distances, or maps are any better.
>>>> The latter details are what I really look for in a model.
>>>>
>>>> On the other hand, there has been good evidence through the years that
>>>> there is useful information in the data beyond an I/sigI of 2 or an
>>>> Rmeas> 100% but getting people to use this data has been a hard slog. The
>>>> reason for this reluctance is that the R values of the resulting models
>>>> are higher. Of course they are higher! That does not mean the models
>>>> are of poorer quality, only that data with lower signal/noise has been
>>>> used that was discarded in the models you used to develop your "gut feeling"
>>>> for the meaning of R.
>>>>
>>>> When you change your criteria for selecting data you have to discard
>>>> your old notions about the acceptable values of empirical quality measures.
>>>> You either have to normalize your measure, as Phil Jeffrey recommends, by
>>>> ensuring that you calculate your R's with the same reflections, or by
>>>> making objective measures of map quality.
>>>>
>>>> Dale Tronrud
>>>>
>>>> P.S. It is entirely possible that refining a model to a very optimistic
>>>> resolution cutoff and calculating the map to a lower resolution might be
>>>> better than throwing out the data altogether.
>>>>
>>>> On 5/1/2012 10:34 AM, Kendall Nettles wrote:
>>>>> I have seen dramatic improvements in maps and behavior during refinement following use of the UCLA anisotropy server in two different cases. For one of them the Rfree went from 33% to 28%. I don't think it would have been publishable otherwise.
>>>>> Kendall
>>>>>
>>>>> On May 1, 2012, at 11:10 AM, Bryan Lepore wrote:
>>>>>
>>>>>> On Mon, Apr 30, 2012 at 4:22 AM, Phil Evans<pre(a)mrc-lmb.cam.ac.uk> wrote:
>>>>>>> Are anisotropic cutoff desirable?
>>>>>> is there a peer-reviewed publication - perhaps from Acta
>>>>>> Crystallographica - which describes precisely why scaling or
>>>>>> refinement programs are inadequate to ameliorate the problem of
>>>>>> anisotropy, and argues why the method applied in Strong, et. al. 2006
>>>>>> satisfies this need?
>>>>>>
>>>>>> -Bryan
>> _______________________________________________
>> phenixbb mailing list
>> phenixbb(a)phenix-online.org
>> http://phenix-online.org/mailman/listinfo/phenixbb
> _______________________________________________
> phenixbb mailing list
> phenixbb(a)phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
13 years, 9 months
[phenixbb] Fwd: Geometry Restraints - Anisotropic truncation
by Kendall Nettles
Hi Tom,
Do you think something like this could be used during refinement to identify the "best" resolution limits? If you have an Rsleep set would Rfree be sufficient for this? I imagine collecting data with a ring of noise and then let the optimal resolution be determined during refinement. My understanding of this is that the modern refinement algorithms can handle some noise in the reflections, but maybe this could be a way to optimize how much signal is needed to contribute in a positive fashion?
Kendall
From: "Terwilliger, Thomas C" <terwilliger(a)lanl.gov<mailto:[email protected]>>
Date: May 3, 2012 10:23:29 AM EDT
To: PHENIX user mailing list <phenixbb(a)phenix-online.org<mailto:[email protected]>>
Subject: Re: [phenixbb] Geometry Restraints - Anisotropic truncation
Reply-To: PHENIX user mailing list <phenixbb(a)phenix-online.org<mailto:[email protected]>>
Hi Kendall,
This could work. You could define a fixed set of test reflections, and never touch these, and never include them in refinement, and always use this fixed set to calculate a free R. Then you could do whatever you want, throw away some work reflections, etc, refine, and evaluate how things are working with the fixed free R set.
All the best,
Tom T
________________________________________
From: phenixbb-bounces(a)phenix-online.org<mailto:[email protected]> [phenixbb-bounces(a)phenix-online.org<mailto:[email protected]>] on behalf of Kendall Nettles [knettles(a)scripps.edu<mailto:[email protected]>]
Sent: Thursday, May 03, 2012 7:05 AM
To: PHENIX user mailing list
Subject: Re: [phenixbb] Geometry Restraints - Anisotropic truncation
Hi Pavel,
Could you use a similar approach to figuring out where to cut your data in general? Could you compare the effects of throwing out reflections in different bins, based on I/sigma, for example, and use this to determine what is truly noise? I might predict that as you throw out "noise" reflections you will see a larger drop in Rfree than from throwing out "signal" reflections, which should converge as you approach the "true" resolution. While we don't use I/sigma exclusively, we do to tend towards cutting most of our data sets at the same i/sigma, around 1.5. It would be great if there was a more scientific approach.
Best,
Kendall
On May 3, 2012, at 7:45 AM, Pavel Afonine wrote:
Hi Kendall,
removing same amount of data randomly gives Rwork/Rfree ~ 30/35%.
Pavel
On 5/3/12 4:13 AM, Kendall Nettles wrote:
Hi Pavel,
What happens if you throw out that many reflections that have signal? Can you take out a random set of the same size?
Best,
Kendall
On May 3, 2012, at 2:41 AM, "Pavel Afonine"<pafonine(a)lbl.gov<mailto:[email protected]>> wrote:
Hi Kendall,
I just did this quick test: calculated R-factors using original and
anisotropy-corrected Mike Sawaya's data (*)
Original:
r_work : 0.3026
r_free : 0.3591
number of reflections: 26944
Truncated:
r_work : 0.2640
r_free : 0.3178
number of reflections: 18176
The difference in R-factors is not too surprising given how many
reflections was removed (about 33%).
Pavel
(*) Note, the data available in PDB is anisotropy corrected. The
original data set was kindly provided to me by the author.
On 5/2/12 5:25 AM, Kendall Nettles wrote:
I didnt think the structure was publishable with Rfree of 33% because I was expecting the reviewers to complain.
We have tested a number of data sets on the UCLA server and it usually doesn't make much difference. I wouldn't expect truncation alone to change Rfree by 5%, and it usually doesn't. The two times I have seen dramatic impacts on the maps ( and Rfree ), the highly anisotrophic sets showed strong waves of difference density as well, which was fixed by throwing out the noise. We have moved to using loose data cutoffs for most structures, but I do think anisotropic truncation can be helpful in rare cases.
Kendall
On May 1, 2012, at 3:07 PM, "Dale Tronrud"<det102(a)uoxray.uoregon.edu<mailto:[email protected]>> wrote:
While philosophically I see no difference between a spherical resolution
cutoff and an elliptical one, a drop in the free R can't be the justification
for the switch. A model cannot be made more "publishable" simply by discarding
data.
We have a whole bunch of empirical guides for judging the quality of this
and that in our field. We determine the resolution limit of a data set (and
imposing a "limit" is another empirical choice made) based on Rmrg, or Rmes,
or Rpim getting too big or I/sigI getting too small and there is no agreement
on how "too big/small" is too "too big/small".
We then have other empirical guides for judging the quality of the models
we produce (e.g. Rwork, Rfree, rmsds of various sorts). Most people seem to
recognize that the these criteria need to be applied differently for different
resolutions. A lower resolution model is allowed a higher Rfree, for example.
Isn't is also true that a model refined to data with a cutoff of I/sigI of
1 would be expected to have a free R higher than a model refined to data with
a cutoff of 2? Surely we cannot say that the decrease in free R that results
from changing the cutoff criteria from 1 to 2 reflects an improved model. It
is the same model after all.
Sometimes this shifting application of empirical criteria enhances the
adoption of new technology. Certainly the TLS parametrization of atomic
motion has been widely accepted because it results in lower working and free
Rs. I've seen it knock 3 to 5 percent off, and while that certainly means
that the model fits the data better, I'm not sure that the quality of the
hydrogen bond distances, van der Waals distances, or maps are any better.
The latter details are what I really look for in a model.
On the other hand, there has been good evidence through the years that
there is useful information in the data beyond an I/sigI of 2 or an
Rmeas> 100% but getting people to use this data has been a hard slog. The
reason for this reluctance is that the R values of the resulting models
are higher. Of course they are higher! That does not mean the models
are of poorer quality, only that data with lower signal/noise has been
used that was discarded in the models you used to develop your "gut feeling"
for the meaning of R.
When you change your criteria for selecting data you have to discard
your old notions about the acceptable values of empirical quality measures.
You either have to normalize your measure, as Phil Jeffrey recommends, by
ensuring that you calculate your R's with the same reflections, or by
making objective measures of map quality.
Dale Tronrud
P.S. It is entirely possible that refining a model to a very optimistic
resolution cutoff and calculating the map to a lower resolution might be
better than throwing out the data altogether.
On 5/1/2012 10:34 AM, Kendall Nettles wrote:
I have seen dramatic improvements in maps and behavior during refinement following use of the UCLA anisotropy server in two different cases. For one of them the Rfree went from 33% to 28%. I don't think it would have been publishable otherwise.
Kendall
On May 1, 2012, at 11:10 AM, Bryan Lepore wrote:
On Mon, Apr 30, 2012 at 4:22 AM, Phil Evans<pre(a)mrc-lmb.cam.ac.uk<mailto:[email protected]>> wrote:
Are anisotropic cutoff desirable?
is there a peer-reviewed publication - perhaps from Acta
Crystallographica - which describes precisely why scaling or
refinement programs are inadequate to ameliorate the problem of
anisotropy, and argues why the method applied in Strong, et. al. 2006
satisfies this need?
-Bryan
_______________________________________________
phenixbb mailing list
phenixbb(a)phenix-online.org<mailto:[email protected]>
http://phenix-online.org/mailman/listinfo/phenixbb
_______________________________________________
phenixbb mailing list
phenixbb(a)phenix-online.org<mailto:[email protected]>
http://phenix-online.org/mailman/listinfo/phenixbb
_______________________________________________
phenixbb mailing list
phenixbb(a)phenix-online.org<mailto:[email protected]>
http://phenix-online.org/mailman/listinfo/phenixbb
_______________________________________________
phenixbb mailing list
phenixbb(a)phenix-online.org<mailto:[email protected]>
http://phenix-online.org/mailman/listinfo/phenixbb
_______________________________________________
phenixbb mailing list
phenixbb(a)phenix-online.org<mailto:[email protected]>
http://phenix-online.org/mailman/listinfo/phenixbb
13 years, 9 months
Re: [cctbxbb] Boost Python 1.56
by richard.gildea＠diamond.ac.uk
I've made a start on transcribing this document here:
https://github.com/cctbx/cctbx_project/wiki/cctbx-Developer-Guidance
It probably still needs cleaning up a bit (e.g. I couldn't figure out quickly how to do 3-level list nesting, 1.a.i etc.) and updating to reflect current practice (e.g. git not svn).
Cheers,
Richard
Dr Richard Gildea
Data Analysis Scientist
Tel: +441235 77 8078
Diamond Light Source Ltd.
Diamond House
Harwell Science & Innovation Campus
Didcot
Oxfordshire
OX11 0DE
________________________________________
From: cctbxbb-bounces(a)phenix-online.org [cctbxbb-bounces(a)phenix-online.org] on behalf of Pavel Afonine [pafonine(a)lbl.gov]
Sent: 06 April 2017 09:07
To: cctbx mailing list
Subject: Re: [cctbxbb] Boost Python 1.56
Hi Graeme,
hm.. this is a good question. We've been through back-and-forth
iterations of editing this file and I think the latest I have is from
Paul. But I can't find a non-PDF version of it. Paul: do you have an
editable version of this file?
Thanks,
Pavel
On 4/6/17 13:45, Graeme.Winter(a)diamond.ac.uk wrote:
> Hi Pavel
>
> These all seem sensible
>
> If you have the original non pdf document it may be easier to transcribe this over..
>
> I also note that it lacks the actual detail on how to run tests! However would be happy to add this once on wiki
>
> Best wishes Graeme
>
> On 6 Apr 2017, at 04:00, Pavel Afonine <pafonine(a)lbl.gov<mailto:[email protected]>> wrote:
>
> Not sure if that answers your questions but once upon a time we here at Berkeley tried to write a some sort of document that was supposed to answer questions like this. Attached. By no means it is complete, up-to-date, etc, but it might be worth reading for anyone who contributes to cctbx. (Even not sure if I'm sending the latest version).
> Unfortunately, nobody bothered to put it in some central place.
>
> Pavel
>
> On 4/6/17 10:51, James Holton wrote:
> Hey Billy,
>
> On a related note. How do I run these regression tests before committing something into Git? Is there a document on dials regression testing I can't find?
>
> -James
>
> On Apr 5, 2017, at 3:38 PM, Billy Poon <bkpoon(a)lbl.gov<mailto:[email protected]>> wrote:
>
> Hi all,
>
> I tested Boost 1.56 on our buildbot servers and got some new test failures with
>
> cctbx_project/scitbx/array_family/boost_python/tst_flex.py
> cctbx_project/scitbx/random/tests/tst_random.py
>
> The full log for CentOS 6 can be found at
>
> http://cci-vm-6.lbl.gov:8010/builders/phenix-nightly-intel-linux-2.6-x86_64…
>
> It looks like the errors are related to random number generation. For a given seed, would the sequence of numbers change when Boost is changed? I did a diff between Boost 1.56 and the current Boost and could not see any changes that immediately stood out as being related to random numbers.
>
> Are these tests failing for others as well?
>
> --
> Billy K. Poon
> Research Scientist, Molecular Biophysics and Integrated Bioimaging
> Lawrence Berkeley National Laboratory
> 1 Cyclotron Road, M/S 33R0345
> Berkeley, CA 94720
> Tel: (510) 486-5709
> Fax: (510) 486-5909
> Web: https://phenix-online.org<https://phenix-online.org/>
>
> On Wed, Apr 5, 2017 at 8:12 AM, Charles Ballard <charles.ballard(a)stfc.ac.uk<mailto:[email protected]>> wrote:
> FYI, we (CCP4) have been using 1.56 for building cctbx/phaser/dials for the last while with no issues. Don't know about 1.60, but 1.59 causes issues with the boost python make_getter and make_setter (initialisation of none const reference if the passed type is a temporary).
>
> Charles
>
> On 3 Apr 2017, at 14:31, Luc Bourhis wrote:
>
> Hi all,
>
> everybody seemed to agree but then it was proposed to move straight to Boost 1.60, and this caused troubles. Could we consider again to move to at least 1.56? As far as I can tell, this does not cause any issue and as stated one year ago, it would help me and Olex 2.
>
> Thanks,
>
> Luc
>
> On 10 Feb 2016, at 15:17, Nicholas Sauter <nksauter(a)lbl.gov<mailto:[email protected]>> wrote:
>
> Nigel, Billy & Aaron,
>
> I completely endorse this move to Boost 1.56. Can we update our build?
>
> Nick
>
> Nicholas K. Sauter, Ph. D.
> Computer Staff Scientist, Molecular Biophysics and Integrated Bioimaging Division
> Lawrence Berkeley National Laboratory
> 1 Cyclotron Rd., Bldg. 33R0345
> Berkeley, CA 94720
> (510) 486-5713<tel:%28510%29%20486-5713>
>
> On Wed, Feb 10, 2016 at 2:41 PM, Luc Bourhis <luc_j_bourhis(a)mac.com<mailto:[email protected]>> wrote:
> Hi,
>
> I have improvements to the smtbx on their way to be committed which require Boost version 1.56. This is related to Boost.Threads, whose support I re-activated a few months ago on Nick’s request. I need the function boost::thread::physical_concurrency which returns the number of physical cores on the machine, as opposed to virtual cores when hyperthreading is enabled (which it is by default on any Intel machine). That function is not available in Boost 1.55 which is the version currently used in the nightly tests: it appeared in 1.56.
>
> So, would it be possible to move to Boost 1.56? Otherwise, I will need to backport that function. Not too difficult but not thrilling.
>
> Best wishes,
>
> Luc
>
>
> _______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org<mailto:[email protected]>
> http://phenix-online.org/mailman/listinfo/cctbxbb
>
> _______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org<mailto:[email protected]>
> http://phenix-online.org/mailman/listinfo/cctbxbb
>
> _______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org<mailto:[email protected]>
> http://phenix-online.org/mailman/listinfo/cctbxbb
>
>
> _______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org<mailto:[email protected]>
> http://phenix-online.org/mailman/listinfo/cctbxbb
>
>
> _______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org<mailto:[email protected]>
> http://phenix-online.org/mailman/listinfo/cctbxbb
>
>
>
> _______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org<mailto:[email protected]>
> http://phenix-online.org/mailman/listinfo/cctbxbb
>
>
> <cctbx-developer-guidance-08-2015.pdf>_______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org<mailto:[email protected]>
> http://phenix-online.org/mailman/listinfo/cctbxbb
>
>
_______________________________________________
cctbxbb mailing list
cctbxbb(a)phenix-online.org
http://phenix-online.org/mailman/listinfo/cctbxbb
--
This e-mail and any attachments may contain confidential, copyright and or privileged material, and are for the use of the intended addressee only. If you are not the intended addressee or an authorised recipient of the addressee please notify us of receipt by returning the e-mail and do not use, copy, retain, distribute or disclose the information in or attached to the e-mail.
Any opinions expressed within this e-mail are those of the individual and not necessarily of Diamond Light Source Ltd.
Diamond Light Source Ltd. cannot guarantee that this e-mail or any attachments are free from viruses and we cannot accept liability for any damage which you may sustain as a result of software viruses which may be transmitted in or with the message.
Diamond Light Source Limited (company no. 4375679). Registered in England and Wales with its registered office at Diamond House, Harwell Science and Innovation Campus, Didcot, Oxfordshire, OX11 0DE, United Kingdom
8 years, 10 months
Re: [phenixbb] Using the Same Test Set in AutoBuild and Phenix.Refine
by Pavel Afonine
Hi Dale,
this tells me that phenix.refine does not see the free-r flags in
exptl_fobs_freeR_flags.mtz file. I'm not sure how Autobuild managed to
run phenix.refine in this case.
Tom is not available until Jan 21. Meanwhile I can try to debug it
myself. To do so I need to reproduce this problem. Could you please send
me the model and data and the exact commands (for autobuild and
phenix.reifne) you used (send it to my mailbox and not to bb)? If the
model / data are confidential feel free to manipulate them somehow
(randomize the data, remove half of the model, ...) but make sure the
program still fail.
Also, I noticed that you are using an old version of PHENIX (I replaced
"main.high_resolution=" with "xray_data.high_resolution=" a while ago).
Since we keep improving things and many bugs were fixed since that time,
it could be a good idea to try to run this with the latest version of
PHENIX (I can do this if you send me the data).
Thanks!
Pavel.
On 1/2/2008 4:47 PM, Dale Tronrud wrote:
> Thomas C. Terwilliger wrote:
>
>> Hi Dale,
>>
>> Can you try something else:
>>
>> phenix.refine AutoBuild_run_12_/overall_best.pdb \
>> refinement.input.xray_data.file_name=\
>> AutoBuild_run_12/exptl_fobs_freeR_flags.mtz \
>> refinement.main.high_resolution=2.2 refinement.main.low_resolution=20 \
>> /usr/users/dale/geometry/chromophores/bcl_tnt.cif
>>
>>
>> This differs from your run only by substituting
>>
>> AutoBuild_run_12/exptl_fobs_freeR_flags.mtz
>>
>> for your 2 refinement data files. This is the exact file that is used in
>> refinement by AutoBuild.
>>
>>
>
> I tried this command, very similar to yours:
>
> phenix.refine AutoBuild_run_12_/overall_best.pdb \
> refinement.input.xray_data.file_name=AutoBuild_run_12_/exptl_fobs_phases_freeR_flags.mtz \
> refinement.main.high_resolution=2.2 refinement.main.low_resolution=20 \
> /usr/users/dale/geometry/chromophores/bcl_tnt.cif output.prefix=junk2
>
> The final output was:
>
> F-obs:
> AutoBuild_run_12_/exptl_fobs_phases_freeR_flags.mtz:FP,SIGFP
>
> If previously used R-free flags are available run this command again
> with the name of the file containing the original flags as an
> additional input. If the structure was never refined before, or if the
> original R-free flags are unrecoverable, run this command again with
> the additional definition:
>
> refinement.main.generate_r_free_flags=True
>
> If the structure was refined previously using different R-free flags,
> the values for R-free will become meaningful only after many cycles of
> refinement.
>
> Sorry: Please try again.
>
> The output from mtz.dump for your .mtz is
>
> Processing: AutoBuild_run_12_/exptl_fobs_phases_freeR_flags.mtz
> Title: Resolve mtz file.
> Space group symbol from file: P
> Space group number from file: 212
> Space group from matrices: P 43 3 2 (No. 212)
> Point group symbol from file: 43
> Number of crystals: 2
> Number of Miller indices: 38159
> Resolution range: 75.6238 2.14896
> History:
> From resolve_huge, 27/12/07 15:07:56
> Crystal 1:
> Name: HKL_base
> Project: HKL_base
> Id: 0
> Unit cell: (169.1, 169.1, 169.1, 90, 90, 90)
> Number of datasets: 1
> Dataset 1:
> Name: HKL_base
> Id: 0
> Wavelength: 0
> Number of columns: 0
> Crystal 2:
> Name: allen-
> Project: FMO-ct
> Id: 2
> Unit cell: (169.1, 169.1, 169.1, 90, 90, 90)
> Number of datasets: 1
> Dataset 1:
> Name: 1
> Id: 1
> Wavelength: 0
> Number of columns: 9
> label #valid %valid min max type
> H 38159 100.00% 0.00 38.00 H: index h,k,l
> K 38159 100.00% 2.00 78.00 H: index h,k,l
> L 38159 100.00% 0.00 55.00 H: index h,k,l
> FP 38159 100.00% 32.00 15171.00 F: amplitude
> SIGFP 38159 100.00% 23.00 1716.00 Q: standard deviation
> PHIM 38159 100.00% -90.00 45.00 P: phase angle in degrees
> FOMM 38159 100.00% 0.00 0.00 W: weight (of some sort)
> FreeR_flag 38159 100.00% 0.00 19.00 I: integer
> FC 38159 100.00% 0.00 0.00 F: amplitude
>
> Dale Tronrud
>
>
>> I agree that you should be able to use your original data file instead. A
>> possible reason why this has failed is that the original data file has a
>> couple reflections for which there is no data...and which were tossed by
>> AutoBuild before creating exptl_fobs_freeR_flags.mtz . Two files that
>> differ only in reflections with no data will still give different
>> checksums, I think.
>>
>> All the best,
>> Tom T
>>
>>
>>> Hi Dale,
>>>
>>>
>>>>> 1) Why you specify reflection MTZ file twice in phenix.refine script?
>>>>>
>>>>>
>>>>>
>>>> I put the mtz in twice because if I put it in once phenix.refine
>>>> complains that I have no free R flags. It seems to want one file with
>>>> the amplitudes and another with the flags. Since I have both in the
>>>> same file I put that file on the line twice and phenix.refine finds
>>>> everything it needs.
>>>>
>>>>
>>> phenix.refine looks for free-R flags in your main data file
>>> (1M50-2.mtz). Optionally you can provide a separate file containing
>>> free-R flags (I have to write about this in the manual). However, if
>>> your 1M50-2.mtz contains free-R flags then you don't need to give it
>>> twice. So clearly something is wrong at this step and we need to find
>>> out what is wrong before doing anything else. Could you send the result
>>> of the command "phenix.mtz.dump 1M50-2.mtz" to see what's inside of your
>>> data file? Or I can debug it myself if you send me the data and model.
>>>
>>>
>>>> If the MD5 hash of the test set depends on the resolution then
>>>> certainly
>>>> I could be in trouble.
>>>>
>>> No. It must always use the original files before any processing.
>>>
>>>
>>>> Does the resolution limit affect the MD5 hash of the test set?
>>>>
>>>>
>>> No. If it does then it is a very bad bug. I will play with this myself
>>> later tonight.
>>>
>>>
>>>>> 3) Does this work:
>>>>>
>>>>> (...)
>>>>>
>>>> I'll try these but it will take a bit of time.
>>>>
>>>>
>>> Don't run it until completion. Just make sure it passed through the
>>> processing step.
>>>
>>> Pavel.
>>>
>>> _______________________________________________
>>> phenixbb mailing list
>>> phenixbb(a)phenix-online.org
>>> http://www.phenix-online.org/mailman/listinfo/phenixbb
>>>
>>>
>> _______________________________________________
>> phenixbb mailing list
>> phenixbb(a)phenix-online.org
>> http://www.phenix-online.org/mailman/listinfo/phenixbb
>>
> _______________________________________________
> phenixbb mailing list
> phenixbb(a)phenix-online.org
> http://www.phenix-online.org/mailman/listinfo/phenixbb
>
18 years, 1 month
Re: [phenixbb] Validation of structure with modified residue
by Xavier Brazzolotto
I recall now how I used to add the glycans before the Carbohydrate Module in Coot.
I have to make the cif file with the putative O atom of SER OG and remove it before making the bond.
I will try that, it will certainly correct my geometry issue but I am still not sure for the final file and validation.
Fingers crossed...
> Le 21 avr. 2022 à 17:09, Xavier Brazzolotto <xbrazzolotto(a)gmail.com> a écrit :
>
> After some careful inspection.
> The geometry on the C atom of the ligand is weird, I don’t get something close to tetrahedral (or similar).
> Probably some angles are missing or I did something wrong with the ligand cif file.
> Not fixed yet
>
>> Le 21 avr. 2022 à 13:39, Xavier Brazzolotto <xbrazzolotto(a)gmail.com <mailto:[email protected]>> a écrit :
>>
>> I’ve re-processed the structure separating the SER residue from the ligand part. Now I have independent ligand.
>> In the « Custom Geometry Restraints » I’ve defined the bond between OG and the carbon atom of the ligand and I’ve defined the angles (I’ve used the values from the previously determined eLBOW run off the SER-bound ligand complex), saved the restraints and launched the refinement. At a first look it was processed correctly and the final cif file has now the whole protein in Chain A.
>>
>> I’ve used prepare PDB deposition using the FASTA sequence of the protein (wonder if I have to provide the ligand CIF file and add more options) and ran phenix.get_pdb_validation : the report looks ok for protein and some other basic ligands (sugars, buffer, glycerol, etc...) but the ligand of interest was not processed (EDS FAILED...). In the PDB file, all these extra ligands are also in Chain A, with water in chain B.
>>
>> If I try the validation through the website (PDBe@EBI) with both cif files from the Refine or the Prepare_PDB_Deposition process, both seem to crash the server as it takes forever without Finalizing...
>>
>> I wonder if I am missing something… Maybe declaration of removal of atoms : HG bound to OG in SER or/and removal of one H from the carbon of the ligand involved in the bond ?
>>
>> Xavier
>>
>>> Le 21 avr. 2022 à 08:06, Xavier Brazzolotto <xbrazzolotto(a)gmail.com <mailto:[email protected]>> a écrit :
>>>
>>> Thank you for your feedback.
>>>
>>> @Paul, I’ve run the « Prepare model for deposition » with the option modified residue (SLG). Not sure it will change if I change the name as it is already the PDB database, but I will give it another try.
>>>
>>> I think that I will have to describe only the ligand and add some parameters restricting distance and angles between the OG of SER and the ligand, I think this is right way.
>>> @ Nigel, is that what you mean with « details » ? If you have any other « tips/tricks » they are welcome.
>>>
>>> Best
>>> Xavier
>>>
>>>> Le 21 avr. 2022 à 02:47, Nigel Moriarty <nwmoriarty(a)lbl.gov <mailto:[email protected]>> a écrit :
>>>>
>>>> Xavier
>>>>
>>>> Paul's point is very valid because the "Prepare for Deposition" step is what generates the sequence (which is the crucial point here) for deposition. However, because you have "created" a new amino acid, there will still be issues as highlighted by Pavel. It is a corner case.
>>>>
>>>> One small addition point is that SLG is already taken in the PDB Ligand list. There are tools in Phenix to find an used code.
>>>>
>>>> Can you re-engineer it with SER+ligand? This will solve the problem using the current Phenix version. I can help with the details if needed.
>>>>
>>>> Cheers
>>>>
>>>> Nigel
>>>>
>>>> ---
>>>> Nigel W. Moriarty
>>>> Building 33R0349, Molecular Biophysics and Integrated Bioimaging
>>>> Lawrence Berkeley National Laboratory
>>>> Berkeley, CA 94720-8235
>>>> Phone : 510-486-5709 Email : NWMoriarty(a)LBL.gov <mailto:[email protected]>
>>>> Fax : 510-486-5909 Web : CCI.LBL.gov <http://cci.lbl.gov/>
>>>> ORCID : orcid.org/0000-0001-8857-9464 <https://orcid.org/0000-0001-8857-9464>
>>>>
>>>>
>>>> On Wed, Apr 20, 2022 at 5:02 PM Paul Adams <pdadams(a)lbl.gov <mailto:[email protected]>> wrote:
>>>>
>>>> Please also remember that you need to run “Prepare model for PDB deposition” (in the GUI under "PDB Deposition") on the mmCIF file you get from phenix.refine. This provides important information that is required for the deposition at the PDB.
>>>>
>>>>> On Apr 20, 2022, at 1:58 PM, Xavier Brazzolotto <xbrazzolotto(a)gmail.com <mailto:[email protected]>> wrote:
>>>>>
>>>>> Dear Phenix users,
>>>>>
>>>>> I don’t know if my problem is related to Phenix but for information I’m running Phenix 1.20.1-4487 under MacOS 12.3.1.
>>>>>
>>>>> I’ve finalized a structure where a ligand covalently modified the protein.
>>>>>
>>>>> I’ve generated the modified residue (named SLG for serine modified by ligand). For this I’ve generated the molecules in SMILES and used eLBOW to generate the restraints. Then I’ve modified the cif file defining the molecule as a L-peptide and replacing the atom names of the Serine part (CA, CB, OG, C, O, N, and OXT)
>>>>> In coot (from CCP4 : 0.9.6 EL), I’ve used the modified cif file and it allowed merging of the modified residue into the polypeptide chain as expected and further refinements went without any issue in Phenix (providing the modified cif file of course). Everything seems well interpreted. So far so good.
>>>>>
>>>>> However, now I would like to validate the structure and both Phenix validation tool and the PDB web server do not accept the final cif file.
>>>>>
>>>>> Checking this file I’ve noticed that the protein seems split into 3 pieces (chain A, first residue up to the one before the modified residue; chain B the modified residue by itself described as HETATM and chain C the rest of the polypeptide up to the C-ter).
>>>>> The PDB file presents only one chain A for the whole protein with the modified residue...
>>>>>
>>>>> I don’t know if this is an issue with Phenix generating this final cif file in this specific case or if I need to modify this final file by hand ?
>>>>>
>>>>> Any help is welcome.
>>>>> Thanks
>>>>>
>>>>> Xavier
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> phenixbb mailing list
>>>>> phenixbb(a)phenix-online.org <mailto:[email protected]>
>>>>> http://phenix-online.org/mailman/listinfo/phenixbb <http://phenix-online.org/mailman/listinfo/phenixbb>
>>>>> Unsubscribe: phenixbb-leave(a)phenix-online.org <mailto:[email protected]>
>>>> --
>>>> Paul Adams (he/him/his)
>>>> Associate Laboratory Director for Biosciences, LBL (https://biosciences.lbl.gov/leadership/ <https://biosciences.lbl.gov/leadership/>)
>>>> Principal Investigator, Computational Crystallography Initiative, LBL (https://cci.lbl.gov <https://cci.lbl.gov/>)
>>>> Vice President for Technology, the Joint BioEnergy Institute (http://www.jbei.org <http://www.jbei.org/>)
>>>> Principal Investigator, ALS-ENABLE, Advanced Light Source (https://als-enable.lbl.gov <https://als-enable.lbl.gov/>)
>>>> Division Deputy for Biosciences, Advanced Light Source (https://als.lbl.gov <https://als.lbl.gov/>)
>>>> Laboratory Research Manager, ENIGMA Science Focus Area (http://enigma.lbl.gov <http://enigma.lbl.gov/>)
>>>> Adjunct Professor, Department of Bioengineering, UC Berkeley (http://bioeng.berkeley.edu <http://bioeng.berkeley.edu/>)
>>>> Member of the Graduate Group in Comparative Biochemistry, UC Berkeley (http://compbiochem.berkeley.edu <http://compbiochem.berkeley.edu/>)
>>>>
>>>> Building 33, Room 250
>>>> Building 978, Room 4126
>>>> Building 977, Room 268
>>>> Tel: 1-510-486-4225
>>>> http://cci.lbl.gov/paul <http://cci.lbl.gov/paul>
>>>> ORCID: 0000-0001-9333-8219
>>>>
>>>> Lawrence Berkeley Laboratory
>>>> 1 Cyclotron Road
>>>> BLDG 33R0345
>>>> Berkeley, CA 94720, USA.
>>>>
>>>> Executive Assistant: Michael Espinosa [ MEEspinosa(a)lbl.gov <mailto:[email protected]> ] [ 1-510-333-6788 ]
>>>> Phenix Consortium: Ashley Dawn [ AshleyDawn(a)lbl.gov <mailto:[email protected]> ][ 1-510-486-5455 ]
>>>>
>>>> --
>>>>
>>>> _______________________________________________
>>>> phenixbb mailing list
>>>> phenixbb(a)phenix-online.org <mailto:[email protected]>
>>>> http://phenix-online.org/mailman/listinfo/phenixbb <http://phenix-online.org/mailman/listinfo/phenixbb>
>>>> Unsubscribe: phenixbb-leave(a)phenix-online.org <mailto:[email protected]>
>>
>
> _______________________________________________
> phenixbb mailing list
> phenixbb(a)phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave(a)phenix-online.org
3 years, 9 months
[phenixbb] Phenix 1.21rc1-5015 Refinement issue on Apple M2
by Xavier Brazzolotto
Dear Phenix users,
Following the slow Molecular Replacement issue in the last stable version 1.20.1-4487 on Apple M2, I’ve tried the last Nightly (1.21rc-5015).
Under the GUI, the MR went OK (as fast as before), but when I try to launch a Refinement (Rigid Body) after the MR, Phenix goes into an error :
"Multiple equally suitable arrays of observed xray data found. »
If I launch the previous version (1.20.1-4487) I am still able to launch the rigid body refinement from the MR results without any issues.
I includes the log file if this helps.
Cheers
Xavier
> Le 4 juil. 2023 à 23:17, Randy John Read <rjr27(a)cam.ac.uk> a écrit :
>
> Fantastic! Sorry it took so long to get my hands on an ARM-based Mac since the first reports of problems!
>
> ----
> Randy J. Read
>
>> On 4 Jul 2023, at 21:02, Luca Jovine <luca.jovine(a)ki.se> wrote:
>>
>> Thank you for the info Randy,
>>
>> I confirm that in the last available nightly build (1.21rc1-5015) the issue is clearly fixed, resulting in a >70-fold speed increase compared to Phaser-MR from 1.21rc1-5008. For two sample jobs using intensity data on a M2 MacBook Pro:
>>
>> 1.21rc1-5008:
>> ------------------
>> Job 1: CPU Time: 0 days 0 hrs 37 mins 23.01 secs ( 2243.01 secs)
>> Job 2: CPU Time: 0 days 0 hrs 31 mins 10.79 secs ( 1870.79 secs)
>>
>> 1.21rc1-5015:
>> -------------------
>> Job 1: CPU Time: 0 days 0 hrs 0 mins 31.13 secs ( 31.13 secs)
>> Job 2: CPU Time: 0 days 0 hrs 0 mins 25.44 secs ( 25.44 secs)
>>
>> ...thanks for the fix!!
>>
>> Luca
>>
>> -----Original Message-----
>> From: Randy John Read <rjr27(a)cam.ac.uk <mailto:[email protected]>>
>> Date: Tuesday, 4 July 2023 at 17:37
>> To: Xavier Brazzolotto <xbrazzolotto(a)gmail.com <mailto:[email protected]>>
>> Cc: PHENIX user mailing list <phenixbb(a)phenix-online.org <mailto:[email protected]>>, Luca Jovine <luca.jovine(a)ki.se <mailto:[email protected]>>
>> Subject: Re: [phenixbb] Discrepancy between Phenix GUI and command line for MR
>>
>>
>> Hi,
>>
>>
>> Thanks for sending the log files!
>>
>>
>> The jobs turn out not actually to be identical. The GUI automatically chose to use the intensity data (which is what we prefer for use in Phaser) whereas your job run from a script is using amplitude data. The issue I alluded to earlier occurs only for intensity data, because the analysis of those data involves applying different equations, which use a special function (tgamma from the Boost library). For some reason I don’t understand, when the Intel version of the tgamma algorithm is computed using the Rosetta functionality to run it on an ARM processor, it’s much much slower than other calculations.
>>
>>
>> Just last week (right after I finally got an M2 MacBook Pro), we tracked this down and replaced the calls to Boost tgamma with alternative code, and that problem should exist any more. You can use it already in Phenix by getting a recent nightly build, and I’ve asked the CCP4 people to compile a new version of Phaser and release it as an update to CCP4 as well.
>>
>>
>> Best wishes,
>>
>>
>> Randy
>>
>>
>>> On 4 Jul 2023, at 12:05, Xavier Brazzolotto <xbrazzolotto(a)gmail.com <mailto:[email protected]>> wrote:
>>>
>>> For information
>>>
>>> Apple M2 running Ventura 13.4.1 with 64 Go memory
>>> Phenix 1.20.1-4487 (Intel one).
>>>
>>> I’ve run MR of the same dataset (2.15A - I422) with the same model both with the command line and through the GUI.
>>>
>>> Command line (phenix.phaser) : 48 secs.
>>> GUI (Phaser-MR simple one component interface): 18 mins !
>>>
>>> In copy the two log files if this helps
>>>
>>>
>>>
>>>>> Le 4 juil. 2023 à 12:54, Luca Jovine <luca.jovine(a)ki.se <mailto:[email protected]>> a écrit :
>>>>
>>>> Hi Xavier and Randy, I'm also experiencing the same on a M2 Mac!
>>>> -Luca
>>>>
>>>> -----Original Message-----
>>>> From: <phenixbb-bounces(a)phenix-online.org <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>> on behalf of Xavier Brazzolotto <xbrazzolotto(a)gmail.com <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>>
>>>> Date: Tuesday, 4 July 2023 at 12:38
>>>> To: Randy John Read <rjr27(a)cam.ac.uk <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>>
>>>> Cc: PHENIX user mailing list <phenixbb(a)phenix-online.org <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>>
>>>> Subject: Re: [phenixbb] Discrepancy between Phenix GUI and command line for MR
>>>>
>>>>
>>>> Hi Randy,
>>>>
>>>>
>>>> Indeed I’m running Phenix on a brand new M2 Mac.
>>>> I will benchmark the two processes (GUI vs command line) and post them here.
>>>>
>>>>
>>>>> Le 4 juil. 2023 à 12:32, Randy John Read <rjr27(a)cam.ac.uk <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>> a écrit :
>>>>>
>>>>> Hi Xavier,
>>>>>
>>>>> We haven’t noticed that, or at least any effect is small enough not to stand out. There shouldn’t be a lot of overhead in communicating with the GUI (i.e. updating the terse log output and the graphs) but if there is we should look into it and see if we can do something about it.
>>>>>
>>>>> Could you tell me how much longer (say, in percentage terms) a job takes when you run it through the GUI compared to running the same job outside the GUI on the same computer? Also, it’s possible the architecture matters so could you say which type of computer and operating system you’re using? If it’s a Mac, is it one with an Intel processor or an ARM (M1 or M2) processor? (By the way, we finally managed to track down and fix an issue that cause Phaser to run really slowly on an M1 or M2 Mac when using the version compiled for Intel, once I got my hands on a new Mac.)
>>>>>
>>>>> Best wishes,
>>>>>
>>>>> Randy
>>>>>
>>>>>> On 4 Jul 2023, at 10:44, Xavier Brazzolotto <xbrazzolotto(a)gmail.com <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>> wrote:
>>>>>>
>>>>>> Dear Phenix users
>>>>>>
>>>>>> I’ve noticed that molecular replacement was clearly slower while running from the GUI compared to using the command line (phenix.phaser).
>>>>>>
>>>>>> Did you also observe such behavior?
>>>>>>
>>>>>> Best
>>>>>> Xavier
>>>>>> _______________________________________________
>>>>>> phenixbb mailing list
>>>>>> phenixbb(a)phenix-online.org <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>
>>>>>> http://phenix-online.org/mailman/listinfo/phenixbb <http://phenix-online.org/mailman/listinfo/phenixbb> <http://phenix-online.org/mailman/listinfo/phenixbb> <http://phenix-online.org/mailman/listinfo/phenixbb;>
>>>>>> Unsubscribe: phenixbb-leave(a)phenix-online.org <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>
>>>>>
>>>>>
>>>>> -----
>>>>> Randy J. Read
>>>>> Department of Haematology, University of Cambridge
>>>>> Cambridge Institute for Medical Research Tel: +44 1223 336500
>>>>> The Keith Peters Building
>>>>> Hills Road E-mail: rjr27(a)cam.ac.uk <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>
>>>>> Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> phenixbb mailing list
>>>> phenixbb(a)phenix-online.org <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>
>>>> http://phenix-online.org/mailman/listinfo/phenixbb <http://phenix-online.org/mailman/listinfo/phenixbb> <http://phenix-online.org/mailman/listinfo/phenixbb> <http://phenix-online.org/mailman/listinfo/phenixbb;>
>>>> Unsubscribe: phenixbb-leave(a)phenix-online.org <mailto:[email protected]> <mailto:[email protected] <mailto:[email protected]>>
>>>>
>>>>
>>>>
>>>> När du skickar e-post till Karolinska Institutet (KI) innebär detta att KI kommer att behandla dina personuppgifter. Här finns information om hur KI behandlar personuppgifter<https://ki.se/medarbetare/integritetsskyddspolicy> <https://ki.se/medarbetare/integritetsskyddspolicy;>.
>>>>
>>>>
>>>> Sending email to Karolinska Institutet (KI) will result in KI processing your personal data. You can read more about KI’s processing of personal data here<https://ki.se/en/staff/data-protection-policy> <https://ki.se/en/staff/data-protection-policy>>.
>>>
>>> <command_line_PHASER.log><GUI_phaser.log>
>>
>>
>> -----
>> Randy J. Read
>> Department of Haematology, University of Cambridge
>> Cambridge Institute for Medical Research Tel: +44 1223 336500
>> The Keith Peters Building
>> Hills Road E-mail: rjr27(a)cam.ac.uk <mailto:[email protected]>
>> Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk
>>
>>
>>
>>
>>
2 years, 7 months
Re: [cctbxbb] some thoughts on cctbx and pip
by Gergely Katona
Dear Billy,
This sounds very promising and exciting. I am not sure if cctbx is
already functional as a conda package in anaconda3 (Linux) or this is
still work in progress. My technical expertise does not allow me to
tell the difference. What I tried:
Fresh install of anaconda3. Adding - cctbx and - conda-forge to
.condarc . Installing pyside2 with conda. Running conda install
conda_dependencies . I get a lot package version conflict, and I
cannot import cctbx or iotbx to anaconda python. Am I following the
right instructions? Or it is too early to expect that cctbx works when
installed through conda?
Best wishes,
Gergely
On Wed, Nov 27, 2019 at 3:56 PM Billy Poon <BKPoon(a)lbl.gov> wrote:
>
> Hi all,
>
> For a brief update, I have submitted a recipe for cctbxlite to conda-forge (https://github.com/conda-forge/staged-recipes/pull/10021) and support for Python 3.7 and 3.8 is being added (https://github.com/cctbx/cctbx_project/pull/409). With some fixes for Windows (https://github.com/cctbx/cctbx_project/pull/416), all platforms (macOS, linux, and Windows) can build for Python 2.7, 3.6, 3.7, and 3.8. Some additional changes will be needed to get Windows to work with Python 3 and for tests to pass with Boost 1.70.0. That will enable the conda-forge recipe to build for all platforms and for all supported versions of Python.
>
> Currently, the conda-forge recipe will install into the conda python and cctbx imports can be done without sourcing the environment scripts. It looks like a lot of the environment variables being set in the dispatchers can be removed since the Python files and C++ extensions are in the right places. I'll update the libtbx_env file so that commands that load the environment can work correctly.
>
> --
> Billy K. Poon
> Research Scientist, Molecular Biophysics and Integrated Bioimaging
> Lawrence Berkeley National Laboratory
> 1 Cyclotron Road, M/S 33R0345
> Berkeley, CA 94720
> Tel: (510) 486-5709
> Fax: (510) 486-5909
> Web: https://phenix-online.org
>
>
> On Sun, Aug 25, 2019 at 2:33 PM Tristan Croll <tic20(a)cam.ac.uk> wrote:
>>
>> Hi Luc,
>>
>> That sounds promising. From there, I’d need to work out how to make a fully-packaged installer (basically a modified wheel file) for the ChimeraX ToolShed - the aim is for the end user to not have to worry about any of this. That adds a couple of complications - e.g. $LIBTBX_BUILD would need to be set dynamically before first import - but doesn’t seem insurmountable.
>>
>> Thanks,
>>
>> Tristan
>>
>>
>>
>> Tristan Croll
>> Research Fellow
>> Cambridge Institute for Medical Research
>> University of Cambridge CB2 0XY
>>
>>
>>
>>
>> > On 25 Aug 2019, at 18:31, Luc Bourhis <luc_j_bourhis(a)mac.com> wrote:
>> >
>> > Hi Tristan,
>> >
>> > cctbx could be built to use your ChimeraX python, now that cctbx is moving to Python 3. The option —with-python is there for that with the bootstrap script. The specific environment setup boil down to setting two environment variable LIBTBX_BUILD and either LD_LIBRARY_PATH on Linux, PATH on Win32, or DYLIB_LIBRARY_PATH on MacOS. If you work within a framework such as ChimeraX, that should not be difficult to ensure those two variables are set.
>> >
>> > Best wishes,
>> >
>> > Luc
>> >
>> >
>> >> On 23 Aug 2019, at 19:02, Tristan Croll <tic20(a)cam.ac.uk> wrote:
>> >>
>> >> To add my two cents on this: probably the second-most common question I've had about ISOLDE's implementation is, "why didn't you use CCTBX?". The honest answer to that is, "I didn't know how."
>> >>
>> >> Still don't, really - although the current developments are rather promising. The problem I've faced is that CCTBX was designed as its own self-contained Python (2.7, until very recently) environment, with its own interpreter and a lot of very specific environment setup. Meanwhile I'm developing ISOLDE in ChimeraX, which is *also* its own self-contained Python (3.7) environment. To plug one into the other in that form... well, I don't think I'm a good enough programmer to really know where to start.
>> >>
>> >> The move to Conda and a more modular CCTBX architecture should make a lot more possible in that direction. Pip would be even better for me personally (ChimeraX can install directly from the PyPI, but doesn't interact with Conda) - but I understand pretty well the substantial challenge that would amount to (not least being that the PyPI imposes a limit - around 100MB from memory? - on the size of an individual package).
>> >>
>> >> Best regards,
>> >>
>> >> Tristan
>> >>
>> >>> On 2019-08-23 09:28, Luc Bourhis wrote:
>> >>> Hi Graeme,
>> >>> Yes, I know. But “black" is a program doing a very particular task
>> >>> (code formatting from the top of my head). Requiring to use a wrapper
>> >>> for python itself is another level. But ok, I think I am mellowing to
>> >>> the idea after all! Talking with people around me, and extrapolating,
>> >>> I would bet that, right now, a great majority of people interested by
>> >>> cctbx in pip have already used the cctbx, so they know about the
>> >>> Python wrapper, and they would not be too sanguine about that. My
>> >>> concern is for the future, when pip will be the first time some people
>> >>> use cctbx. Big fat warning notices on PyPI page and a better error
>> >>> message when cctbx fails because LIBTBX_BUILD is not set would be
>> >>> needed but that could be all right.
>> >>> If we do a pip installer, we should aim at a minimal install: cctbx,
>> >>> iotbx and their dependencies, and that’s it.
>> >>> Best wishes,
>> >>> Luc
>> >>>> On 23 Aug 2019, at 07:17, Graeme.Winter(a)Diamond.ac.uk <Graeme.Winter(a)diamond.ac.uk> wrote:
>> >>>> Without discussing the merits of this or whether we _choose_ to make the move to supporting PIP, I am certain it would be _possible_ - many other packages make dispatcher scripts when you pip install them e.g.
>> >>>> Silver-Surfer rescale_f2 :) $ which black; cat $(which black)
>> >>>> /Library/Frameworks/Python.framework/Versions/3.6/bin/black
>> >>>> #!/Library/Frameworks/Python.framework/Versions/3.6/bin/python3.6
>> >>>> # -*- coding: utf-8 -*-
>> >>>> import re
>> >>>> import sys
>> >>>> from black import main
>> >>>> if __name__ == '__main__':
>> >>>> sys.argv[0] = re.sub(r'(-script\.pyw?|\.exe)?$', '', sys.argv[0])
>> >>>> sys.exit(main())
>> >>>> So we _could_ work around the absence of LIBTBX_BUILD etc. in the system. Whether or not we elect to do the work is a different question, and it seems clear that here are very mixed opinions on this.
>> >>>> Best wishes Graeme
>> >>>> On 23 Aug 2019, at 01:21, Luc Bourhis <luc_j_bourhis(a)mac.com<mailto:[email protected]>> wrote:
>> >>>> Hi,
>> >>>> Even if we managed to ship our the boost dynamic libraries with pip, it would still not be pip-like, as we would still need our python wrappers to set LIBTBX_BUILD and LD_LIBRARY_PATH. Normal pip packages work with the standard python exe. LD_LIBRARY_PATH, we could get around that by changing the way we compile, using -Wl,-R, which is the runtime equivalent of build time -L. That’s a significant change that would need to be tested. But there is no way around setting LIBTBX_BUILD right now. Leaving that to the user is horrible. Perhaps there is a way to hack libtbx/env_config.py so that we can hardwire LIBTBX_BUILD in there when pip installs?
>> >>>> Best wishes,
>> >>>> Luc
>> >>>> On 16 Aug 2019, at 22:47, Luc Bourhis <luc_j_bourhis(a)mac.com<mailto:[email protected]>> wrote:
>> >>>> Hi,
>> >>>> I did look into that many years ago, and even toyed with building a pip installer. What stopped me is the exact conclusion you reached too: the user would not have the pip experience he expects. You are right that it is a lot of effort but is it worth it? Considering that remark, I don’t think so. Now, Conda was created specifically to go beyond pip pure-python-only support. Since cctbx has garnered support for Conda, the best avenue imho is to go the extra length to have a package on Anaconda.org<http://anaconda.org/>, and then to advertise it hard to every potential user out there.
>> >>>> Best wishes,
>> >>>> Luc
>> >>>> On 16 Aug 2019, at 21:45, Aaron Brewster <asbrewster(a)lbl.gov<mailto:[email protected]>> wrote:
>> >>>> Hi, to avoid clouding Dorothee's documentation email thread, which I think is a highly useful enterprise, here's some thoughts about putting cctbx into pip. Pip doesn't install non-python dependencies well. I don't think boost is available as a package on pip (at least the package version we use). wxPython4 isn't portable through pip (https://wiki.wxpython.org/How%20to%20install%20wxPython#Installing_wxPython…). MPI libraries are system dependent. If cctbx were a pure python package, pip would be fine, but cctbx is not.
>> >>>> All that said, we could build a manylinux1 version of cctbx and upload it to PyPi (I'm just learning about this). For a pip package to be portable (which is a requirement for cctbx), it needs to conform to PEP513, the manylinux1 standard (https://www.python.org/dev/peps/pep-0513/). For example, numpy is built according to this standard (see https://pypi.org/project/numpy/#files, where you'll see the manylinux1 wheel). Note, the manylinux1 standard is built with Centos 5.11 which we no longer support.
>> >>>> There is also a manylinux2010 standard, which is based on Centos 6 (https://www.python.org/dev/peps/pep-0571/). This is likely a more attainable target (note though by default C++11 is not supported on Centos 6).
>> >>>> If we built a manylinuxX version of cctbx and uploaded it to PyPi, the user would need all the non-python dependencies. There's no way to specify these in pip. For example, cctbx requires boost 1.63 or better. The user will need to have it in a place their python can find it, or we could package it ourselves and supply it, similar to how the pip h5py package now comes with an hd5f library, or how the pip numpy package includes an openblas library. We'd have to do the same for any packages we depend on that aren't on pip using the manylinux standards, such as wxPython4.
>> >>>> Further, we need to think about how dials and other cctbx-based packages interact. If pip install cctbx is set up, how does pip install dials work, such that any dials shared libraries can find the cctbx libraries? Can shared libraries from one pip package link against libraries in another pip package? Would each package need to supply its own boost? Possibly this is well understood in the pip field, but not by me :)
>> >>>> Finally, there's the option of providing a source pip package. This would require the full compiler toolchain for any given platform (macOS, linux, windows). These are likely available for developers, but not for general users.
>> >>>> Anyway, these are some of the obstacles. Not saying it isn't possible, it's just a lot of effort.
>> >>>> Thanks,
>> >>>> -Aaron
>> >>>> _______________________________________________
>> >>>> cctbxbb mailing list
>> >>>> cctbxbb(a)phenix-online.org<mailto:[email protected]>
>> >>>> http://phenix-online.org/mailman/listinfo/cctbxbb
>> >>>> _______________________________________________
>> >>>> cctbxbb mailing list
>> >>>> cctbxbb(a)phenix-online.org<mailto:[email protected]>
>> >>>> http://phenix-online.org/mailman/listinfo/cctbxbb
>> >>>> _______________________________________________
>> >>>> cctbxbb mailing list
>> >>>> cctbxbb(a)phenix-online.org<mailto:[email protected]>
>> >>>> http://phenix-online.org/mailman/listinfo/cctbxbb
>> >>>> --
>> >>>> This e-mail and any attachments may contain confidential, copyright and or privileged material, and are for the use of the intended addressee only. If you are not the intended addressee or an authorised recipient of the addressee please notify us of receipt by returning the e-mail and do not use, copy, retain, distribute or disclose the information in or attached to the e-mail.
>> >>>> Any opinions expressed within this e-mail are those of the individual and not necessarily of Diamond Light Source Ltd.
>> >>>> Diamond Light Source Ltd. cannot guarantee that this e-mail or any attachments are free from viruses and we cannot accept liability for any damage which you may sustain as a result of software viruses which may be transmitted in or with the message.
>> >>>> Diamond Light Source Limited (company no. 4375679). Registered in England and Wales with its registered office at Diamond House, Harwell Science and Innovation Campus, Didcot, Oxfordshire, OX11 0DE, United Kingdom
>> >>>> _______________________________________________
>> >>>> cctbxbb mailing list
>> >>>> cctbxbb(a)phenix-online.org
>> >>>> http://phenix-online.org/mailman/listinfo/cctbxbb
>> >>> _______________________________________________
>> >>> cctbxbb mailing list
>> >>> cctbxbb(a)phenix-online.org
>> >>> http://phenix-online.org/mailman/listinfo/cctbxbb
>> >>
>> >>
>> >> _______________________________________________
>> >> cctbxbb mailing list
>> >> cctbxbb(a)phenix-online.org
>> >> http://phenix-online.org/mailman/listinfo/cctbxbb
>> >
>> >
>> > _______________________________________________
>> > cctbxbb mailing list
>> > cctbxbb(a)phenix-online.org
>> > http://phenix-online.org/mailman/listinfo/cctbxbb
>>
>>
>> _______________________________________________
>> cctbxbb mailing list
>> cctbxbb(a)phenix-online.org
>> http://phenix-online.org/mailman/listinfo/cctbxbb
>
> _______________________________________________
> cctbxbb mailing list
> cctbxbb(a)phenix-online.org
> http://phenix-online.org/mailman/listinfo/cctbxbb
--
Gergely Katona, PhD
Associate Professor
Department of Chemistry and Molecular Biology, University of Gothenburg
Box 462, 40530 Göteborg, Sweden
Tel: +46-31-786-3959 / M: +46-70-912-3309 / Fax: +46-31-786-3910
Web: http://katonalab.eu, Email: gergely.katona(a)gu.se
6 years, 1 month