Boulder Laboratory for 3-Dimensional Electron Microscopy of Cells
REFINEMATCH(1) REFINEMATCH(1)
NAME
refinematch - Solve for a refined match between two tomograms
SYNOPSIS
refinematch [options] input_file [output_file]
DESCRIPTION
Refinematch will solve for a general 3-dimensional linear transformation
to align two volumes to each other. It performs multiple linear
regression on the displacements between the volumes determined at a matrix
of positions. The displacements must be contained in a file with the
following form:
Number of displacements
One line for each displacement consisting of the X, Y, and Z
coordinates in the first volume, then the displacements in X, Y
and Z involved in moving from the first to the second volume
The program will automatically eliminate "outliers", patch displacements
that are likely to be incorrect because they are so extreme, when compared
to the rest of the displacements. This elimination is conservative, but
if for some reason it operates incorrectly, you can control the parameters
of elimination or stop the elimination from occurring. The program will
eliminate up to 10% of the patches. If more than this number are bad,
either get a new set of patches that do not approach so close to the edge
of the volume, make a model in the tomogram with contours enclosing the
patches to use, eliminate the bad ones from the file by hand, or use
Findwarp to eliminate a whole row or column of patches. When the mean
residual exceeds a value that you specify, the program will exit with an
error.
A model specifying which patches to include in the fit can be quite
simple, consisting of just a single contour enclosing the region where
patches are generally good. This contour can be drawn in any Z plane of
the flipped tomogram. However, if the good region changes through the
depth of the tomogram, you can draw contours at several Z levels. If you
have two layers of patches, draw two contours, one near the top and one
near the bottom of the tomogram; if you have three layers, add another
contour in the middle, etc. For a given patch, the program will find the
contour at the nearest Z level and use that one to determine whether to
include the patch.
If there is only one layer of patches in one dimension, there is
insufficient information to solve for the full transformation, so the
program will solve for only two of the three columns of the transformation
matrix. This typically occurs in the Y dimension, in which case the
second column of the matrix is fixed at 0, 1, 0.
Refinematch uses the PIP package for input (see the manual page for pip)
but can still take input interactively for compatibility with old versions
of Matchorwarp. The following options can be specified either as
command line arguments (with the -) or one per line in a command file or
parameter file (without the -):
-patch OR -PatchFile File name
Name of input file with positions and displacements. If this option is
not entered, the first non-option argument will be used for the input file
name.
-output OR -OutputFile File name
Optional output file for the refining transformation. If this option is
not entered, the second non-option argument (if any) will be used for the
output file name.
-region OR -RegionModel File name
Model file with contours enclosing the patches to be included in the fit
-volume OR -VolumeOrSizeXYZ File name
Either the name of the file or the X, Y, and Z dimensions of the volume
being matched to.
-residual OR -ResidualPatchOutput File name
Output file for positions, displacements, and residual values. If the
patch input file has correlation coefficients, they will be replaced by
the residuals.
-reduced OR -ReducedVectorOutput File name
Output file for residual vectors from the fit, which represent the vectors
remaining after removing the linear component of the vector field. These
are referred to as reduced vectors. If the input patch file has
correlation coefficients, they will be passed into this output file;
otherwise the residual values will be placed into the file.
-limit OR -MeanResidualLimit Floating point
Limiting value for the mean residual; above this value, the program will
exit with an error.
-maxfrac OR -MaxFractionToDrop Floating point
Maximum fraction of patches to drop from the fit by outlier elimination.
Enter 0 to for no outlier elimination. The default is 0.1.
-minresid OR -MinResidualToDrop Floating point
The minimum residual for outlier elimination; patches with residuals
smaller than this value will be retained no matter how extreme they are
relative to the other patches. The default is 0.5.
-prob OR -CriterionProbabilities Two floats
Two probabilities controlling outlier elimination: a criterion probability
for a patch to be evaluated as an outlier (default 0.01), and a criterion
probability for a patch to be eliminated regardless of the distribution of
extreme values (default 0.002).
-param OR -ParameterFile Parameter file
Read parameter entries as keyword-value pairs from a parameter file.
-help OR -usage
Print help output
-StandardInput
Read parameter entries from standard input.
If the program is started with no command line arguments, it takes
interactive input with the following entries:
Name of file with positions and displacements
Either the file name, or the X, Y, and Z dimensions of the volume being
matched to.
Name of an IMOD model file with contours enclosing the patches to
be included in the fit, or Return to use all patches.
Limiting value for the mean residual; above this value, the program
will exit with an error.
Name of output file in which to place the transformation, or Return
for no output to a file
HISTORY
Written by David Mastronarde, 1995
12/24/98: added outlier elimination; 6/6/99: added error exit
8/21/06: converted to PIP, made it handle both orientations of volume
better, changed outlier output to be a summary as in Findwarp,
added options for controlling outlier elimination and getting
residual output.