On 10/19/2012 08:59 PM, Quentin Delettre wrote:

Dear Phenix users,

I am new to phenix and the Durandal tool, i would like to know if Durandal is an appropriate way to cluster many homology models of a protein.

If and only if the input PDBs follow the requirement from the README file: --- The alignment expected by the Durandal software is 1 to 1. For example, if 2 structures A and B are to be optimally superposed before computing the CARMSD, both PDB files (A.pdb and B.pdb) must have same numbers of alpha carbons. The first alpha carbon in A.pdb will be matched to the first alpha carbon in B.pdb. The second alpha carbon in A.pdb will be matched to the second alpha carbon in B.pdb. And so on until the end of both files. If it is not the case, please edit all your PDB files in advance. ---

And if so, what is the syntax of the output ? The man page does not provide any information beside inputs commands.

I will try to update the README with the following: Use the -v option to get more readable cluster listings. Then, the output looks like this: --- pole position centers(5): ./pdbs/02.pdb ./pdbs/03.pdb ./pdbs/05.pdb ./pdbs/07.pdb ./pdbs/09.pdb members(3): ./pdbs/01.pdb ./pdbs/03.pdb ./pdbs/07.pdb members(3): ./pdbs/02.pdb ./pdbs/05.pdb ./pdbs/09.pdb members(2): ./pdbs/00.pdb ./pdbs/06.pdb members(2): ./pdbs/04.pdb ./pdbs/08.pdb --- A cluster is listed after a 'members(N):' line. The first PDB of the cluster is the cluster center, other PDBs are cluster members. Clusters are listed in a biggest-first (most members) order. The 'pole position centers(N):' and following lines indicate that there were other possible biggest clusters, only their centers are listed. Regards, Francois.

Thanks. _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb