Obtaining Output

The Fitted Structure

The fitted mobile structure may be written to a file in PDB format using the WRITE command:

   WRITE fitted.pdb

If the first character of the filename is a pipe character ($\mid$), then the results will be piped into the specified program. For example:

   WRITE |less

will cause the coordinates to be displayed on the screen using the less pager program.

The reference set may also be written:

   WRITE REFERENCE ref_fitted.pdb

(only the three letters `REF' of the REFERENCE parameter are required). This is only useful if the CENTRE command has been used (see below).

Centering the Coordinates

By default, the mobile structure is moved to the coordinate frame of the reference set. If the CENTRE (or CENTER) command is given then the centre of geometry of the fitted coordinates will be located at the origin.

If a residue has been set as the centre of fitting using SETCENTRE (see Section 9.6) then that residue will be moved to the origin when the CENTRE command is used.

If only two structures are fitted then the WRITE REFERENCE command must be used to write the reference set in the origin-centred coordinate frame. If multiple structures are fitted and written using MWRITE then the reference set will be written automatically.

Details of the Fitting

More details about the fitting may be obtained by using the MATRIX command. This displays the centres of geometry, the rotation matrix and the translation vector which is the vector between the centres of geometry. Thus to superimpose the mobile structure onto the reference structure using these data, you should translate the mobile set to the origin, apply the rotation matrix, translate back to the original centre of geometry and finally apply the translation vector.

Note that the rotation matrix is not orthogonal and cannot therefore be used to extract Euler angles. This is a result of the fitting method used.

The NFITTED command displays the number of atom pairs which were fitted in the last fitting operation. Note that this will not be the number of residues fitted unless you are only fitting one atom type per residue (typically C$\alpha$ atoms).

By-residue RMS Deviation

The RESIDUE command is used to obtain a by-residue RMS deviation on the currently specified RMS atoms in the currently specified RMS zone. If no RATOMS and RZONE commands have been used, the atoms and zones used for the fitting will be used.

The RESIDUE command may be followed by an optional filename parameter in which case output is directed to the specified file. If the file cannot be opened or a filename is not specified, output appears on the screen. If the first character of the filename is a pipe character ($\mid$), then the results will be piped into the specified program. For example:

   RESIDUE |less

will cause the results to be displayed on the screen using the less pager program.

If the distance cutoff is set then residues fully outside the distance cutoff are flagged with ** and residues partially outside the distance cutoff are flagged with * (see Section 9.7)

The related command, PAIRDIST prints the pairwise distances between equivalent atom pairs in the reference and mobile structures. PAIRDIST has the same syntax as RESIDUE. If the distance cutoff is set then residues outside the distance cutoff are flagged with *.

Outputting Fit Zones As An Alignment

As of ProFit V3.0, it is possible to output the equivalenced regions found by iterative fitting as an alignment using the PRINTALIGN command. The default output is a (user-friendly) pairwise alignment with the reference and mobile sequences printed as pairs of 60-character wide lines. The optional FASTA and PIR parameters set the printout to (machine-friendly) FASTA or PIR formatting for the chain names and sequences.

Alignments can be exported to a text file using the PRINTALIGN command. ProFit V3.0 can read PIR formatted files for assigning zones.

Note: For a set of fit zones to be converted into an alignment, the fitting zones must occur sequentially along the protein chain. Additionally, the fit zones cannot overlap. In other words, to obtain a sequence alignment the fitting zones must be in sequence.