next up previous contents index
Next: COMPARE compare Up: Comparison and searching of Previous: ALIGN_CONSENSUS consensus   Contents   Index

SUPERPOSE -- superpose MODEL2 on MODEL given alignment

ALIGN_CODES = $\langle{\tt string:0}\rangle$ 'all' codes of proteins in the alignment
FIT = $\langle{\tt logical:1}\rangle$ on whether to superpose
SUPERPOSE_REFINE = $\langle{\tt logical:1}\rangle$ off whether to refine the superposition
RMS_CUTOFFS = $\langle{\tt real:11}\rangle$ 3.5 3.5 60 60 15 60 60 60 60 60 60 only the first element is used for calculating the cutoff RMS and DRMS measures
REFERENCE_ATOM = $\langle{\tt string:1}\rangle$ '' reference atom name in SUPERPOSE
REFERENCE_DISTANCE = $\langle{\tt real:1}\rangle$ 3.5 cutoff for selecting reference positions in SUPERPOSE
SWAP_ATOMS_IN_RES = $\langle{\tt string:1}\rangle$ '' minimize RMS by swapping atoms in these residues (1 char code: 'DEFHLNQRVY')

Requirements:
MODEL & MODEL2 [& alignment]

Description:
This command superposes MODEL2 on MODEL, without changing the alignment.

If an alignment is in memory, it is used to obtain the equivalent atoms. MODEL must be the first sequence in the alignment, MODEL2 must be the second sequence in the alignment. The equivalent atoms are those selected atoms (set 1) of the MODEL that have equivalently named atoms in MODEL2; the atom equivalences are defined in library $ATMEQV_LIB. Use the PICK_ATOMS command to select the desired atoms for superposition. By default, all atoms are selected. If there is no alignment, a 1:1 correspondence between the residues is assumed.

No fitting is done if FIT = off.

The ALIGN_CODES variable is used only for output, not in calculations.

The RMS_CUTOFFS[1] element is the cutoff used in calculating the cutoff RMS deviations; i.e., those position and distance RMS deviations that are defined on the equivalent atoms which are less than RMS_CUTOFFS[1] angstroms away from each other (as superposed using all aligned positions) and those equivalent distances which are less than RMS_CUTOFFS[1] angstroms different from each other, respectively.

If SUPERPOSE_REFINE is on the refinement of the superposition is done by repeating the fitting with only those aligned pairs of atoms that are within RMS_CUTOFFS[1] of each other until there is no change in the number of equivalent positions. This refinement can only remove compared positions, not add them like ALIGN3D can do. This is useful for comparing equivalent parts of two structures with a fixed alignment but omitting divergent parts from the superposition and RMS deviation calculation; e.g., comparing a model with the X-ray structure.

If SUPERPOSE_REFINE is off and REFERENCE_ATOM is non-blank, only those pairs of equivalently named selected atoms from aligned residues are superposed that come from residues whose REFERENCE_ATOM atoms are closer than REFERENCE_DISTANCE to each other.

When MODEL and MODEL2 have exactly the same atoms in the same order, one can set SWAP_ATOMS_IN_RES to any combination of single character amino acid residue codes in DEFHLNQRVY. Certain atoms (see below) in the specified sidechains of MODEL2 are then swapped to minimize their RMS deviation relative to MODEL. The labelling resulting in the lowest RMS deviation is retained. The following swaps are attempted:

Residue Swap(s)
D OD1, OD2
E OE1, OE2
F CD1, CD2
  CE1, CE2
H ND1, CD2
  NE2, CE1
N OD1, ND2
Q OE1, NE2
R NH1, NH2
V CG1, CG2
Y CD1, CD2
  CE1, CE2

Example:


# Example for: SUPERPOSE

# This will use a given alignment to superpose Calpha atoms of 
# one structure (2ctx) on the other (1fas).

READ_MODEL  FILE = '1fas'
READ_MODEL2 FILE = '2ctx'
SET ALIGN_CODES = '1fas' '2ctx'
READ_ALIGNMENT FILE = 'toxin.ali'
PICK_ATOMS PICK_ATOMS_SET = 1, ATOM_TYPES = 'CA'
SUPERPOSE
WRITE_MODEL2 FILE = '2ctx.fit'

Example:


# Example for: ALIGN3D, SUPERPOSE

# This will align 3D structures of two proteins:

SET OUTPUT_CONTROL = 1 1 1 1 1

# First example: read sequences from a sequence file:
READ_ALIGNMENT FILE = 'toxin.ali', ALIGN_CODES = '1fas' '2ctx'
ALIGN GAP_PENALTIES_1D= -600 -400
ALIGN3D GAP_PENALTIES_3D= 0 4.0
WRITE_ALIGNMENT FILE = 'toxin-str.ali'

# Second example: read sequences from PDB files to eliminate the 
# need for the toxin.ali sequence file:
READ_MODEL FILE = '1fas'
SEQUENCE_TO_ALI ATOM_FILES = '1fas', ALIGN_CODES = '1fas'
READ_MODEL FILE = '2ctx'
SEQUENCE_TO_ALI ADD_SEQUENCE = on, ATOM_FILES = ATOM_FILES '2ctx', ;
                ALIGN_CODES = ALIGN_CODES '2ctx'
ALIGN GAP_PENALTIES_1D= -600 -400
ALIGN3D GAP_PENALTIES_3D=  0 2.0
WRITE_ALIGNMENT FILE = 'toxin-str.ali'

# And now superpose the two structures using current alignment to get 
# various RMS's:
READ_MODEL  FILE = '1fas'
PICK_ATOMS ATOM_TYPES = 'CA'
READ_MODEL2 FILE = '2ctx'
SUPERPOSE FIT_ATOMS = 'CA'

Example:


# This script illustrates the use of the SWAP_ATOMS_IN_RES
# argument to the SUPERPOSE command:

# Need to make sure that the topologies of the two molecules
# superposed are exactly the same:

READ_TOPOLOGY FILE = '$(LIB)/top_heav.lib'
READ_PARAMETERS FILE = '$(LIB)/par.lib'
READ_MODEL FILE = '../tutorial-model/1fdx.atm'
SEQUENCE_TO_ALI ALIGN_CODES = '1fdx', ATOM_FILES = FILE
SEQUENCE_TO_ALI ADD_SEQUENCE = on, ALIGN_CODES = ALIGN_CODES '1fdx', ATOM_FILES = ATOM_FILES FILE
GENERATE_TOPOLOGY SEQUENCE = '1fdx'
TRANSFER_XYZ
BUILD_MODEL INITIALIZE_XYZ = off

# READ_MODEL2 FILE = '../tutorial-model/1fdx.B99990002'
READ_MODEL2 FILE = './1fdx.swap.atm'
SET SWAP_ATOMS_IN_RES = ''
SUPERPOSE 
SET SWAP_ATOMS_IN_RES = 'DEFHLNQRVY'
SUPERPOSE FIT = off
SET SWAP_ATOMS_IN_RES = ''
SUPERPOSE FIT = on


next up previous contents index
Next: COMPARE compare Up: Comparison and searching of Previous: ALIGN_CONSENSUS consensus   Contents   Index
Ben Webb 2004-04-20