Hi Matt, Currently each input conformation for morphing can only be one model. Chimera considers each input file to be one model (or more, if the file contains multiple sets of coordinates delimited with MODEL/ENDMDL records).

Without code change, the only way I can think of is to put each entire structure of 24 monomers in a single PDB file. This may be a pain, as you would probably want to give each monomer a unique chain ID so that all residues in the "model" are still uniquely identifiable by residue number/chain ID. Then you would have one file per conformation.

Since I haven't actually tried this, I recommend trying on a subset of 2-3 chains before going to all the trouble of putting all 24 in a file! Maybe the others can come up with a more elegant solution. Best, Elaine ----- Elaine C. Meng, Ph.D. meng@cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html

On Jul 27, 2008, at 8:13 PM, Dougherty, Matthew T. wrote:

I have a complex structure in two conformations: resting and active.

Each structure consists of 24 monomers, corresponding to 24 pdb files.

Each monomer file between conformations matches in terms of atom numbers, chain ID, etc.

I can morph a single monomer between conformations.

I am having no success morphing all the monomers simultaneously between the conformations.

Any suggestions on how to organize my data to overcome this?

thanks, Matt

Hi Matt, At least in my smaller-scale tests, it worked to morph multiple chains in one model to multiple chains in another model. What I tried:

(a) Lin-7/Lin-2: PDB 1zl8 is an NMR structure with A and B chains. It contains multiple models delimited with MODEL/ENDMDL records. I tried morphing from the first to the second to the third model and confirmed that both A and B chains morphed.

(b) insulin: PDB 1aph has A and B chains, PDB 1fub has A-D, where either A,B or C,D could be paired with the first structure. I opened 1aph, opened 1fub and deleted its C,D chains - then morphing worked. Then to confirm different chain IDs are OK, I closed and reopened 1fub, then deleted its A,B chains - again morphing worked.

(c) antibody Fab fragment: since insulin's A and B chains are disulfide-linked, I wanted to try another example where the chains aren't covalently bonded, just in case. PDB 1kem has L,H chains while PDB 1mh5 has L,H but another set A,B. I deleted L,H from 1mh5 and confirmed that it did work to morph between the two structures and that both chains moved.

I believe that atom and residue numbering are not used at all and have morphed between many structures with different numbering, such as homologous proteins with fairly low percent identity.

In your agglomerated files, I believe that although the chain IDs do not need to match between conformation 1 and conformation 2, the chain IDs should be unique in each file. That is, no file should contain more than one chain with the same ID. Other than that, I don't have ideas of what could be amiss with your input.

Do you get any messages from Chimera, or does it happily go about its business of not generating what you want? Best, Elaine ----- Elaine C. Meng, Ph.D. meng@cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html

P.S. I was using last night's build, but I believe there have not been any changes in Morph Conformations for several weeks (should be identical to the last production release).

On Jul 28, 2008, at 9:39 AM, Dougherty, Matthew T. wrote:

we tried the one file approach, did not work out. will try again. does it make any difference about the chain id or atom number?

Matthew Dougherty 713-433-3849 National Center for Macromolecular Imaging Baylor College of Medicine/Houston Texas USA = = = ====================================================================== = = = ======================================================================