multifiltsetup(1) General Commands Manual multifiltsetup(1)
NAME
multifiltsetup - Test multiple reconstruction filter values on a sub-
volume
SYNOPSIS
multifiltsetup options tilt_com_file_name
DESCRIPTION
Multifiltsetup produces command files for reconstructing a defined
region of a tomogram with a series of values for one of the types of
radial filters in Tilt. Four kinds of filters can be varied, but
only one kind can be changed on any given run.
Two filters providing an alternative form for the ramp function, the
SIRT-like filter and "exact filters", can be varied with:
-fake List of iteration values
-exact List of "object sizes"
The two alternative high-frequency filtering options, the standard
Gaussian filter and the Hamming-like filter, can be varied with:
-cutoff List of cutoffs and/or -falloff List of falloffs
-hamming List of start frequencies
The behavior and requirements when varying the Gaussian filter are com-
plicated because there are two parameters. The rules are these:
1) If both -cutoff and -falloff are entered, either both must have
the same number of frequencies listed, or one must have only a single
value.
2) If -falloff is entered but not -cutoff, a fixed cutoff for all
runs is taken from the "RADIAL" entry in the input command file.
3) If -falloff is entered with multiple values and -cutoff with one
value, that cutoff is used for all of the runs
4) If -cutoff is entered but not -falloff, a fixed ratio of falloff
to cutoff is taken from the ratio of the falloff and cutoff values of
the "RADIAL" entry in the input command file.
5) If -cutoff is entered with multiple values and -falloff with one
value, a fixed ratio of falloff to cutoff is taken from the ratio of
the entered falloff value and the first cutoff.
6) If -cutoff is entered with multiple values and a falloff is not
entered for each, then the falloff values will all be this fixed ratio
times their respective cutoff values.
This scaling will not work if the first cutoff is 0, or if the falloff
is 0.
For example, if -cutoff is entered as "0.1,0.2,0.3", no -falloff is
entered, and the "RADIAL" entry in the file is "0.35 0.035", then the
ratio will be 0.035 / 0.35 = 0.1 and the falloffs will be 0.01, 0.02,
and 0.03. If -falloff is entered with just "0.02", the ratio will be
0.02 / 0.1 = 0.2 and the falloffs will be 0.02, 0.04, and 0.06.
Unless options are entered to specify a subvolume, the default is to
reconstruction the full area with the same thickness.
OPTIONS
Multifiltsetup uses the PIP package for input (see the manual page for
pip). Options can be specified either as command line arguments
(with the -) or one per line in a command file (without the -).
Options can be abbreviated to unique letters; the currently valid
abbreviations for short names are shown in parentheses.
-com (-co) OR -CommandFile File name
Command file for reconstruction, which can be entered with or
without its extension of ".com". If this option is not entered,
a non-option argument is used for the command file.
-fake (-fak) OR -FakeSIRTiterations List of integer ranges
List of iteration numbers for runs with SIRT-like filtering.
Ranges are allowed. An entry for "ExactFilterSize" will be
removed from the command file and the high-frequency filtering
option will be retained.
-exact (-e) OR -ExactObjectSizes List of integer ranges
List of "object sizes" for runs with "exact filters". Ranges
are allowed. An entry for "FakeSIRTiterations" will be will be
removed from the command file and the high-frequency filtering
option will be retained.
-hamming (-ha) OR -HammingLikeStarts Multiple floats
Set of starting frequencies, in reciprocal pixels (0 - 0.5) for
a Hamming-like filter. An entry for "RADIAL" will be removed
from the command file and any options about the initial form of
the filter will be retained.
-cutoffs (-cu) OR -GaussianCutoffs Multiple floats
Set of frequencies for the cutoff of the standard Gaussian fil-
ter, in reciprocal pixels (0 - 0.5). See the rules above for
how the cutoff and falloff vary when this option does not have
the same number of entries as the -falloff option.
-falloffs (-fal) OR -GaussianFalloffs Multiple floats
Set of frequencies for the falloff of the standard Gaussian fil-
ter, in reciprocal pixels (0 - 0.5).
-width (-w) OR -WidthInX Integer
Width of tomograms to build, in unbinned pixels. The default is
to do the whole extent in X. An existing "WIDTH" entry in the
command file will be replaced or removed.
-ysize (-ysi) OR -SizeInY Integer
Size in Y (the Y dimension of the aligned stack) of tomograms to
build, in unbinned pixels. The default is to do the whole
extent in Y. An existing "SLICE" entry in the command file will
be replaced or removed.
-thick (-t) OR -ThicknessInZ Integer
Thickness of tomograms to make, in unbinned pixels. The default
is to use the "THICKNESS" entry in the command file.
-xshift (-x) OR -ShiftInX Floating point
Shift of tomograms in X, in unbinned pixels. For this and the
other shift entries, a positive shift will bring an area at neg-
ative offset to the center. The default is 0 shift; an existing
"SHIFT" entry in the command file will be replaced or removed.
-yshift (-ysh) OR -ShiftInY Integer
Shift of tomograms in Y (the Y dimension of the aligned stack),
in unbinned pixels. The default is 0 shift.
-zshift (-z) OR -ShiftInDepth Floating point
Shift of tomograms in Z (the depth dimension), in unbinned pix-
els. The default is to use the Z shift value from the "SHIFT"
entry in the command file, if any.
-help (-he) OR -usage
Print help output
-StandardInput
Read parameter entries from standard input
FILES
The command files are named "tilt_mulfil-*.com" for an input file
"tilt.com", "tilt_mulfila-*.com" for "tilta.com", etc. For -fake and
-exact options, they produce tomograms "setname_slfinnn.mrc" and "set-
name_efosnnn.mrc", where "setname" is the root name of the data set and
"nnn" is the iteration number or object size, respectively, and all
files in a run will be 2, 3, or 4 digits with leading zeros as needed.
For the -hamming option, the tomograms are "setname_hlfs0.xxx.mrc"
where "0.xxx" is the starting frequency to 3 decimal places. When
varying the Gaussian filter, the tomogram names contain both the cutoff
and falloff as "setname_gfc0.xxx-f0.xxx.mrc".
AUTHOR
David Mastronarde
SEE ALSO
tilt, proceschunks(1)
BUGS
Email bug reports to mast at colorado dot edu.
IMOD 5.2.0 multifiltsetup(1)