AMBER Archive (2000)

Subject: Re: Using CARNAL to compute per-residue RMSD between PDB files

From: Bill Ross (ross_at_cgl.ucsf.edu)
Date: Thu Nov 02 2000 - 15:35:08 CST


The RMS residues question:

I think this is what you want - I should have a pointer to it in
the RMS section if I don't already:

          TABLE tbid { column_list } ;
              At least one column must be specified. Columns are
              printed in their order in the list. Column_list may
              include ids, classes of measurement (e.g. DIST) which
              print in the order declared, MEAS which prints all
              scalar measurements, or ALL which prints everything.
              AXIS ids result in vectors, PLANE ids in normals, and
              GROUP ids default to center of geometry unless
              attributes are specified, such as grpid%cmass. RMS
              ids default to the rms of the group, while
              rmsid%residues and rmsid%atoms give per-residue and
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
              per-atom rms respectively. For per-residue rms, the
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
              group must not have any partially-included residues.
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
              If either per-residue or per-atom options are used,
              the statistics are printed in the summary with the
              residue and atom names.

You can also fit on 1 group and measure on another, which gives
the ability to get rms for partial residues, although tediously.

Bill