Re: [PyMOL] pdbml

2015-10-22 Thread Arthur Zalevsky
Hi Thomas,

thank you very much for help.

Since PDB made pdbml default for large systems (i'm particularly 
interested in ribosomes) it was a kind of pain.
And it seems i have to upgrade my PyMOL and read documentation more 
often: help for fetch in PyMOL from svn shows '{default: cif (default 
was "pdb" up to 1.7.6)}'.

Regards,
Arthur


On 23.10.2015 00:48, Thomas Holder wrote:
> Hi Arthur,
> 
> The script I sent you earlier had one Incentive PyMOL specific line
> and won't work in Open-Source PyMOL unless you delete that line. My
> apologies.
> 
> However, if you are compiling the latest code from SVN you don't need
> the script, I already have incorporated it into the code base! Simply
> load files with .xml or .pdbml extension.
> 
> I have to add that loading 4v50 in PDBML format into PyMOL is a rather
> bad idea. PDBML is utterly bloated, the uncompressed file for 4v50 is
> over 300MB, and the in-memory xml representation will explode into the
> gigabytes. You should really use mmCIF format instead, PyMOL loads
> 4v50.cif in 2 seconds.
> 
> Cheers,
>   Thomas
> 
> On 22 Oct 2015, at 12:38, Arthur Zalevsky  
> wrote:
> 
>> Hi Thomas!
>> 
>> I've tried your parser with pymol 1.4.1 from ubuntu 12.04 repos and
>> freshly compiled 1.7.7.2 from svn with no luck at all. I've tried to
>> load PDB ID 4V50 (full E.coli ribosome). Pymol simply hangs with 100%
>> cpu usage.
>> 
>> From 1.4.1 i've got following traceback (after i left it overnight):
>> 
>> Traceback (most recent call last):
>>   File "/usr/lib/python2.7/dist-packages/pymol/parser.py", line 254, 
>> in
>> parse
>> self.result=apply(layer.kw[0],layer.args,layer.kw_args)
>>   File "/home/domain/silwer/Downloads/load_pdbml.py", line 80, in
>> load_pdbml
>> atom.properties['entity_id'] = child.text or ''
>>   File "/usr/lib/python2.7/dist-packages/chempy/__init__.py", line 59,
>> in __getattr__
>> raise AttributeError(attr)
>> AttributeError: properties
>> 
>> Any help would be appreciated.
>> 
>> 
>> Regards,
>> Arthur
>> 
>> 
>> On 25.09.2015 01:50, Thomas Holder wrote:
>>> Hi Arthur & Jerome,
>>> 
>>> I wrote a basic PDBML importer for PyMOL, see attached script which
>>> adds a "load_pdbml" command. I meant to assess the complexity of such
>>> a project and ended up writing the entire thing. So far, this only
>>> reads atoms and symmetry information. If you find this useful we can
>>> include that in the next PyMOL release.
>>> 
>>> Cheers,
>>>  Thomas
>>> 
>>> 
>>> 
>>> On 23 Sep 2015, at 03:27, Arthur Zalevsky 
>>> wrote:
>>> 
 Sorry for bumping old thread, but is support for pdbml still
 unimplemented? In such case i think it's a good project for our
 students.
 
 Regards,
 Arthur
 
> From: Thomas Holder  - 2013-12-02 19:23:18
> Hi Jerome,
> 
> I think the simple reason is that nobody ever wrote a PDBML parser
> for
> PyMOL. Is that a good reason?
> 
> Cheers,
> Thomas
> 
>> On 28 Nov 2013, at 13:21, Jerome BENOIT  wrote:
>> Hello List,
>> 
>> is there any good reason why PyMOL can not read PDBML data files ?
>> 
>> Cheers,
>> Jerome

--
___
PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net)
Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users
Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net


Re: [PyMOL] pdbml

2015-10-22 Thread Thomas Holder
Hi Arthur,

The script I sent you earlier had one Incentive PyMOL specific line and won't 
work in Open-Source PyMOL unless you delete that line. My apologies.

However, if you are compiling the latest code from SVN you don't need the 
script, I already have incorporated it into the code base! Simply load files 
with .xml or .pdbml extension.

I have to add that loading 4v50 in PDBML format into PyMOL is a rather bad 
idea. PDBML is utterly bloated, the uncompressed file for 4v50 is over 300MB, 
and the in-memory xml representation will explode into the gigabytes. You 
should really use mmCIF format instead, PyMOL loads 4v50.cif in 2 seconds.

Cheers,
  Thomas

On 22 Oct 2015, at 12:38, Arthur Zalevsky  wrote:

> Hi Thomas!
> 
> I've tried your parser with pymol 1.4.1 from ubuntu 12.04 repos and 
> freshly compiled 1.7.7.2 from svn with no luck at all. I've tried to 
> load PDB ID 4V50 (full E.coli ribosome). Pymol simply hangs with 100% 
> cpu usage.
> 
> From 1.4.1 i've got following traceback (after i left it overnight):
> 
> Traceback (most recent call last):
>   File "/usr/lib/python2.7/dist-packages/pymol/parser.py", line 254, in 
> parse
> self.result=apply(layer.kw[0],layer.args,layer.kw_args)
>   File "/home/domain/silwer/Downloads/load_pdbml.py", line 80, in 
> load_pdbml
> atom.properties['entity_id'] = child.text or ''
>   File "/usr/lib/python2.7/dist-packages/chempy/__init__.py", line 59, 
> in __getattr__
> raise AttributeError(attr)
> AttributeError: properties
> 
> Any help would be appreciated.
> 
> 
> Regards,
> Arthur
> 
> 
> On 25.09.2015 01:50, Thomas Holder wrote:
>> Hi Arthur & Jerome,
>> 
>> I wrote a basic PDBML importer for PyMOL, see attached script which
>> adds a "load_pdbml" command. I meant to assess the complexity of such
>> a project and ended up writing the entire thing. So far, this only
>> reads atoms and symmetry information. If you find this useful we can
>> include that in the next PyMOL release.
>> 
>> Cheers,
>>  Thomas
>> 
>> 
>> 
>> On 23 Sep 2015, at 03:27, Arthur Zalevsky  
>> wrote:
>> 
>>> Sorry for bumping old thread, but is support for pdbml still
>>> unimplemented? In such case i think it's a good project for our
>>> students.
>>> 
>>> Regards,
>>> Arthur
>>> 
 From: Thomas Holder  - 2013-12-02 19:23:18
 Hi Jerome,
 
 I think the simple reason is that nobody ever wrote a PDBML parser 
 for
 PyMOL. Is that a good reason?
 
 Cheers,
 Thomas
 
> On 28 Nov 2013, at 13:21, Jerome BENOIT  wrote:
> Hello List,
> 
> is there any good reason why PyMOL can not read PDBML data files ?
> 
> Cheers,
> Jerome

-- 
Thomas Holder
PyMOL Principal Developer
Schrödinger, Inc.


--
___
PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net)
Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users
Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net


Re: [PyMOL] pdbml

2015-10-22 Thread Arthur Zalevsky
Hi Thomas!

I've tried your parser with pymol 1.4.1 from ubuntu 12.04 repos and 
freshly compiled 1.7.7.2 from svn with no luck at all. I've tried to 
load PDB ID 4V50 (full E.coli ribosome). Pymol simply hangs with 100% 
cpu usage.

 From 1.4.1 i've got following traceback (after i left it overnight):

Traceback (most recent call last):
   File "/usr/lib/python2.7/dist-packages/pymol/parser.py", line 254, in 
parse
 self.result=apply(layer.kw[0],layer.args,layer.kw_args)
   File "/home/domain/silwer/Downloads/load_pdbml.py", line 80, in 
load_pdbml
 atom.properties['entity_id'] = child.text or ''
   File "/usr/lib/python2.7/dist-packages/chempy/__init__.py", line 59, 
in __getattr__
 raise AttributeError(attr)
AttributeError: properties

Any help would be appreciated.


Regards,
Arthur


On 25.09.2015 01:50, Thomas Holder wrote:
> Hi Arthur & Jerome,
> 
> I wrote a basic PDBML importer for PyMOL, see attached script which
> adds a "load_pdbml" command. I meant to assess the complexity of such
> a project and ended up writing the entire thing. So far, this only
> reads atoms and symmetry information. If you find this useful we can
> include that in the next PyMOL release.
> 
> Cheers,
>   Thomas
> 
> 
> 
> On 23 Sep 2015, at 03:27, Arthur Zalevsky  
> wrote:
> 
>> Sorry for bumping old thread, but is support for pdbml still
>> unimplemented? In such case i think it's a good project for our
>> students.
>> 
>> Regards,
>> Arthur
>> 
>>> From: Thomas Holder  - 2013-12-02 19:23:18
>>> Hi Jerome,
>>> 
>>> I think the simple reason is that nobody ever wrote a PDBML parser 
>>> for
>>> PyMOL. Is that a good reason?
>>> 
>>> Cheers,
>>> Thomas
>>> 
 On 28 Nov 2013, at 13:21, Jerome BENOIT  wrote:
 Hello List,
 
 is there any good reason why PyMOL can not read PDBML data files ?
 
 Cheers,
 Jerome
> 
> --
> 
> ___
> PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net)
> Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users
> Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net

--
___
PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net)
Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users
Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net