[Chimera-users] Suggestion on biological assembly load

[Thomas Goddard]

Hi Miguel,

   Adding a fetch by id option for PQS biological assembly PDB files 
looks like a good idea.  I'll put it on the requested features list.

   Some of those files reuse atom serial numbers and reuse chain 
identifiers.  For example, satellite tobacco mosaic virus 1a34 which 
involves 60 copies of the asymmetric unit (4 chains) has more than 
200,000 atoms and hundreds of chains.  The PDB file format can't handle 
that correctly.  Still the file displays in Chimera although it can be 
hard to select just a single capsid protein since it shares a chain 
identifier with other subunits.  This is a very small virus.  Bigger 
viruses just don't have PQS entries -- they probably set a size limit. 
So this isn't a good way to handled icosahedral viruses.  But probably 
most other PDB entries will not have size problems.

   Another issue is that PQS has more than one biological assembly for 
some PDB entries.  For example, 1a6v has 3 assemblies, with files called 
1a6v_1.mmol, 1a6v_2.mmol and 1a6v_3.mmol.  I guess it would make sense 
to load all of the assemblies as separate Chimera models in those cases.

   Thanks for the suggestion.

	Tom