[Chimera-users] saving sessions

Giovanni Cardone gcardone at ucsd.edu
Wed May 8 11:13:12 PDT 2013


Hi Tom,

on this regard I think it would be nice to have an separate option, when 
saving a session, to create a self-contained session, which would 
generate a directory containing the session file plus all the data 
attached to that session. In this way you can take that directory and 
move it elsewhere, send it to a collaborator, etc., knowing that you can 
open it without scrambling for the original data (even after years, 
given the great backward compatibility of chimera). That would be also 
useful for archiving purposes (often a session corresponds to a figure 
in a paper). Personally I would prefer the maps out of the session 
files, even if it's true that you can open the session and save the file 
separately (although it's a two-step process).
Just my rambling thoughts.

Giovanni


On 05/08/2013 09:50 AM, Tom Goddard wrote:
> Hi Thomas,
>
>   There is no session saving option that copies density map files to 
> the same directory as the session file.  All other Chimera models 
> including PDB files are included  (not linkied) in the session file. 
>  The reason maps are not put into the session file is because they are 
> sometimes very large, Gbytes in size.  Some day I intend to add an 
> option to include the maps in the session file so you can have a 
> self-contained session if you want. Copying the files could be another 
> approach but it has problems.  What if there is already a map with the 
> same file name in the current directory?  Could modify the copied file 
> name I guess.  Maybe better put all the map files in a new 
> subdirectory.  But are you really going to make copies of all the map 
> files on every session save?  If you only do it sometimes it doesn't 
> solve the problem of moving data to a new computer.  It will be slow 
> to copy the files and it makes many redundant copies which then can 
> become confusing as to whether two files with the same name really are 
> identical.  Hard links could be used to avoid copying the file, but 
> that won't work when a session is on one file system and data on 
> another.  I think the way to handle moving data files to a new 
> computer is to keep all your data including sessions under one 
> directory tree so the whole thing can be easily copied.  If instead 
> the goal is to occassionally make a self-contained session then 
> copying the map files and session file to a new directory would be a 
> nice option similar to embedding all files in the session.  I've noted 
> this in the feature request for embedding the map files.
>
> http://plato.cgl.ucsf.edu/trac/chimera/ticket/9763
>
>   The sphere/axes display idea is one way to see the axes of a 
> displayed model.  The transparent sphere would cause problems in 
> Chimera some of the time because Chimera can only correctly display 
> one transparent model at a time.  The current default is to show a 
> single transparent layer, so any transparent model inside the sphere 
> would disappear.  This is a longstanding and hard problem to fix.  But 
> a mesh could work.  Another idea is simply to depict the x,y,z axes. 
>  I've made a feature request for this.
>
> http://plato.cgl.ucsf.edu/trac/chimera/ticket/12062
>
>   Tom
>
>
>
> On May 8, 2013, at 12:25 AM, "Marlovits,Thomas"  wrote:
>
>> Hi Tom,
>>
>> I do have a question related to saving sessions, which is a great 
>> option btw.
>> Is there a way that one can save the session - let's say in a new 
>> folder but then have all models also saved in there.  For now, as I 
>> understand, it will only save the path/filename of those models.  So, 
>> whenever we switch computers, hard-drives etc ... this can turn out 
>> to be a lengthy process to find those models (we often have many 
>> models open at the same time).
>> Ideally, also fetched files from the pdb would be great to be saved 
>> locally in this new folder.
>>
>> The second question I have is related to turning molecules around 
>> specific x/y/z angles.
>> I'd love to have an option to display a big sphere that encapsulated 
>> the other models displayed.  Ideally, this sphere would look like a 
>> globe with cordinates in there ... if this is displayed at 
>> semi-transparency, one can easily go to a specific position but get 
>> more of an idea to easier naviagte through the 3D space.  Sometimes I 
>> only would like to get a rough estimate about the angles to turn 
>> molecules.  Alternatively a mesh-like grid could do the same job (and 
>> if one could display the numbers of the lattidude/longitude or so 
>> that would be great too).
>>
>> Many tx,
>> -T.
>>
>>
>> -------------------------------------------------------------------------------
>>
>> Thomas C. Marlovits, PhD, M.A.S.
>> The MarlovitsLab @ IMP/IMBA
>> http://thomas.marlovits.myresearchlab.org 
>> <http://thomas.marlovits.myresearchlab.org/>
>>
>> Contact/Address/Directions:
>> http://thomas.marlovits.myresearchlab.org/lab/Contact.html
>>
>
>
>
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://plato.cgl.ucsf.edu/pipermail/chimera-users/attachments/20130508/6d6e99a4/attachment-0001.html>


More information about the Chimera-users mailing list