Hello!
I have redone the basic tutorial successfully with the example given, and would now like to reproduce those same steps with my protein of interest. In my working directory, I have the PDB files for the structures I'm trying to align and the following compare.py file:
------------------------------------------------ env = environ() aln = alignment(env, file='$(LIB)/CHAINS_all.seq', align_codes=('1n75', '1n77')) aln.malign() aln.malign3d() aln.compare_structures() aln.id_table() env.dendrogram() ---------------------------------------------------
I was not sure what $(LIB)/CHAINS_all.seq referred to.
If I need to change that line:
aln = alignment(env, file='$(LIB)/CHAINS_all.seq',
Then what type of file should I use?
This is the error message I am getting:
----------------------------------------------------- File "compare.py", line 3, in ? align_codes=('1n75', '1n77')) File "/usr/local/packages/modeller8v1/modlib/modeller/alignment.py", line 19, in __init__ self.__int_append('alignment.append', io, libs, vars) File "/usr/local/packages/modeller8v1/modlib/modeller/alignment.py", line 68, in __int_append libs=libs.modpt, **vars) File "/usr/local/packages/modeller8v1/modlib/modeller/util/top.py", line 33, in read_alignment return _modeller.read_alignment(aln, io, libs, *args) _modeller.error: read_al_373E> Protein specified in ALIGN_CODES(i) was not found in the alignment file; ALIGN_CODES( 1) = 1n75 ----------------------------------------------------
and my compare.log:
------------------------------------------------- Kind, OS, HostName, Kernel, Processor: 4, Linux pelican 2.6.7-gentoo-r11 i686 Date and time of compilation : 06/20/2005 12:04:54 MODELLER executable type : i386-intel8 Job starting time (YY/MM/DD HH:MM:SS): 2005/09/16 11:37:06.955
read_al_373E> Protein specified in ALIGN_CODES(i) was not found in the alignment file; ALIGN_CODES( 1) = 1n75 --- -----------------------------------------------
Thank you for any suggestions as to how I can compare the structures!
-Charlotte
Hi,
I had two quick questions: I was wondering whether the model returned by MODELLER changes when the template is a backbone trace, in contrast to when it is the full structure with sidechains.
Also, in general, is there any advantage of building the sidechains of the homology model, after running MODELLER, with a tool such as MAXSPROUT (which uses a rotamer library to construct the sidechains)?
Thanks,
--Serkan
Charlotte Habegger-Polomat wrote: > I have redone the basic tutorial successfully with the example given, > and would now like to reproduce those same steps with my protein of > interest. In my working directory, I have the PDB files for the > structures I'm trying to align and the following compare.py file: > > ------------------------------------------------ > env = environ() > aln = alignment(env, file='$(LIB)/CHAINS_all.seq', > align_codes=('1n75', '1n77')) > aln.malign() > aln.malign3d() > aln.compare_structures() > aln.id_table() > env.dendrogram() > --------------------------------------------------- > > > I was not sure what $(LIB)/CHAINS_all.seq referred to.
$(LIB) refers to the directory containing Modeller library files (usually the modlib/ subdirectory wherever you installed Modeller). CHAINS_all.seq is simply a file in PIR alignment format which contains the sequence of every chain in the PDB (as of the last Modeller update).
In your case, you can either use this file directly, since 1n75 and 1n77 are both in there (although you will also need to specify the chain identifier, so it'll be 1n75A and either 1n77A or 1n77B) or you can read the sequences from the PDB files themselves using the alignment.append_model() command: http://salilab.org/modeller/manual/node181.html
Ben Webb, Modeller Caretaker
-
Charlotte Habegger-Polomat -
Mehmet Serkan Apaydin -
Modeller Caretaker