[modeller_usage] multiple protein conformations

Hi,

I have a single sequence and several structures of different conformations of the same protein.  I would 
like to use this information to create a single alignment that incorporates the data from all the 
conformations.  The structures are all of the same protein, so they have the same sequence.  Once I 
have the single alignment, then I can use this alignment to create a model for each conformation.  Also 
note that because of the different conformations, the alignment of the sequence with structure A is not 
(necessarily) the same alignment given if I align the sequence with structue B.
  
My main approach has been to append the structures and the sequence to alignment and then run 
align2d().  The problem I have here is that because all the structures have the same sequence, the 
returned alignment just has all the protein sequences aligned with no gaps and my single sequence is 
just added with all the extra residues overhanging(the sequence is a little longer that the protein 
sequences) the alignment of the proteins.  Is there some way to add a weight to the protein sequences 
so that they don't overpower the single sequence?   Since the protein sequences align perfectly, the 
addition of the single sequence is negligible to the overall alignent.  

I thought this was what the align_what='BLOCK' argument was supposed to do, but I didn't have any 
luck with that either--I tried to put the proteins in one block, and the sequence in another. I've also 
been trying salign() without success. 

I'm new to Modeller and this problem seems sufficiently different from the tutorials and examples in the 
manual to leave me at a loss.  Would someone give me some guidance about how to approach this?

Thank you,
Jim Fonseca