As part of my use of Modeller, I would like to access the secondary structure information for the model. This looks to be achieved using the SSM option to model.write_data, but this produces a file which doesn’t contain the chain identifier.
This could be worked around using the first column (integer index) of the file and knowledge of the chains in the PDB, but it seems awkward to have to do this when the PSA option to write_data is able to include the chain identifier.
Is there an alternative way to access the secondary structures, or could this be fixed in Modeller to make using write_data easier?
Thanks in advance,
On 1/18/19 12:35 AM, dms305 wrote: > As part of my use of Modeller, I would like to access the secondary > structure information for the model. This looks to be achieved using > the SSM option to model.write_data, but this produces a file which > doesn’t contain the chain identifier.
If you're only interested in the secondary structure type, you can read this directly from the biso field - no need to use an intermediate file at all. Something like the following would work:
mdl.write_data(output='SSM') for c in mdl.chains: for r in c.residues: print(c.name, r.num, r.name, int(r.atoms.biso))
If you also want to look at the other columns in the SSM file, you're right - there's no chain ID output and so you'd have to use the residue index instead. I'll add chain IDs in the next Modeller release.
Ben Webb, Modeller Caretaker