mcps
[options] <CCP4 map file>
mcps
will plot a section (or a stack of sections) from a CCP4 map
file using both a (dithered monochrome) grayscale representation and contour
lines. This may be useful in cases where the electron or potential density
distribution has "high valleys" or "low peaks" that make simple contour-line
plots confusing.
The program comes with a (hopefully) sensible set of default values which
should allow you to give it a try with just
mcps my.map
(to plot
the first section of the map), or
mcps -first 3 my.map
to plot
the
3rd
section.
The output (postscript) file is always written to a file whose name is
constructed by appending the suffix ".ps" to the name of the input CCP4 map
file.
mcps
will plot the grayscale gradient starting from 1.0 sigma below
the mean (white), to 3.0 sigma above the mean (black). Contour lines will be
plotted starting from 0.50 sigma below the mean and then every 0.50 sigma.
All contour lines that correspond to density higher that 3.0 sigma above the
mean (and are, thus, on a black background) will be drawn white.
To plot the first section of a CCP4 map file with the name
myfile.map
using both contours and a grayscale representation, give
mcps myfile.map
To plot the 6th section of the map with tick marks every 0.250 fractional
units, give
mcps -first 6 -ticks 0.250 myfile.map
All input to the program is through (case-insensitive) command-line options.
Abbreviated (but uniquely identifiable) forms of the various flags are also
valid.
Where
W1
is the (integer) section to be plotted, or the first of a
stack of sections. The valid range of values for this parameter is defined
by the contents of the map and you can find it from the map header that
mcps
is writing out when it starts (look for the last two numbers on
the line starting with "Start and stop points on ...")
Where
W2
is the (integer) last section for plotting a stack of
sections. See the description in the next keyword for an explanation of what
is the actual calculation performed.
This keyword changes the way the stack of sections is calculated.
Because the expression "stack of sections" may (rightly) create confusion, I
should explain that due to the grayscale representation, the transparency of
the individual sections is lost, and so, what is plotted is not a stack of
sections, but a projection of these sections. How is this projection
calculated ? In the default mode the program will go through all sections,
and for each grid point will keep the maximum value encountered. This avoids
problems with negative regions (which could mask high density features on a
previous or forthcoming sections), but will also increase the noise level.
When the keyword
-sum
is present, the program will calculate
the average value (over all sections) of each grid point, and it is this
"average density map" that will be plotted.
Where
RHO
is the minimum density (of the map file) for the grayscale
gradient. Areas with density below
RHO
will appear "white" in the
final plot (unless, of course, the contrast is inverted with the option
-reverse).
Where
RHO
is the maximum density (of the map file) for the grayscale
gradient. Areas with density greater than
RHO
will appear "black" in the
final plot (unless, of course, the contrast is inverted with the option
-reverse).
Where
RHO
is the density (of the map file) at which the first contour
line will be plotted.
Where
RHO
is the interval of density (of the map file) for plotting
successive (after the first) contour lines.
Where
RHO
is the density (of the map file) at which the contrast of
the contour lines will be reversed (from black to white, or the reverse
depending on the absence or presence of the
-reverse
flag).
If this flag is present, the plotting of contour lines will be switched-off
(leaving only the grayscale representation).
If this flag is present, the contrast of the whole plot will be reversed
(don't you just hate those fancy laser printer toners ?).
Where
VAL
is a scale factor to be applied to the final graph. Please
note that this
does not
change the resolution (see
-resol)
of the plot. The result is that if you use the default
resolution (of 300 dpi) with a
-scale 0.5,
the final plot will have
an effective resolution of 600 dpi. The way to produce a smaller plot
with the same resolution is to simultaneously define
-resol
and
-scale
(for example, the flags "-reso 0.5 -scale 0.5" would produce
a postscipt file with half the size of the default plot but at a constant
resolution of 300dpi).
Where
VAL
is a scale factor to be applied to the resolution of the
final graph. Take care with this keyword : the default resolution for the
program is 300dpi and if you double it (with "-resol 2.0") you may well find
that you need a large amount of physical memory for performing the
calculation [the program will warn you (and pause for a few seconds) if more
than 32 MBytes of memory are about to be requested]. See also second
paragraph in the section
Bugs.
If this flag is present, the orientation of the axes in the final plot will
be swapped, that is, the graph will be drawn with the medium-changing axis
vertical (instead of the default, which has the fast-changing axis running
vertically).
If this flag is present, the plotting of axes and tick marks will be
switched off.
Where
VAL
is the interval in fractional coordinates for drawing tick
marks along the axes.
Some additional (meaningless) numbers are written out during program execution. Not
useful.
mcps
will
not
re-calculate any of map statistics (like mean,
minimum, maximum, rms deviation, etc.). These are all taken from the map
header, and if they are not correct, neither will the plot be.
mcps
will always produce a postscript file which at the default
magnification will have (depending of the unit cell dimensions) a width of 7
inches or a height of 10 inches. Because this whole area is sampled at the
default resolution of 300 dpi, the resulting postscript files (although
bitmapped monochrome) will be quite large (of the order of MBytes) and the
procedure of calculating them is much slower than for normal contour plots.
mcps
produces a bitmapped dithered monochrome image at a default
resolution of 300dpi. The problem with this is, that while the result will look
quite good when printed, the usual pre-viewing methods (even ghostscript)
will produce a rather loosy approximation to it. One of the possible
work-arounds is to actually produce an intermediate file at the correct
resolution, which you then display at a reduced magnification. If, for
example, you have ImageMagick on your machine, you can try something like
display -density 300 -geometry 50% myccp4.map.ps
Nicholas M. Glykos, Department of Molecular Biology and Genetics, Democritus University of Thrace, Alexandroupolis, Greece.
You can get the latest release of the program via
http://utopia.duth.gr/~glykos/