{version = 1.09; (* of malopt.p 1996 March 22} (* begin module describe.malopt *) (* name malopt: 2d plot of optimal alignments synopsis malopt(optalign: in, maloptp: in, xyin: out, output: out) files optalign: output of malign maloptp: parameters to control this program. Lines: 1: doclasses: integer number of classes to process 2: colorbound: highest distance of the color scale 3: xrectangle, yrectangle: size of rectangles of colors. Values of 1 and 1 give nice squares that just fit together. Values of 0.9 and 0.9 squares that don't touch each other. (This would allow more precise location of particluar pieces of data.) xyin: input to xyplo. Columns: 1: x class 2: y class 3: x H, uncertainty in bits 4: y H, uncertainty in bits 5: x rectangle width 6: y rectangle height 7: distance from x class to y class 8: hue for color computed from the distance 9: saturation for color 10: brightness for color A color scale is put on the left side of the graph, running from 0 to colorbound. The lower right triangle of the graph gives the distances according to these colors. output: messages to the user description How can we look at all the classes produced by malign? We can first normalize each alignment vector by subtracting the first value from all other values. Then we can ask what the distance is between all classes. These data are put into xyin for plotting by the xyplo program. The distances are computed and reported to xyin. The distances are also linearly transformed to give a color scale. The user can define the upper bound of the color scale, to be able to look at details for smaller distances. Distances larger than this "colorbound" are set to black. examples documentation see also malign.p, maloptp, xyplo.p, xyplop.malopt author Thomas Dana Schneider bugs technical notes See xyplo for notes on color. *) (* end module describe.malopt *) {This manual page was created by makman 1.44}{created by htmlink 1.55}