cba (version 0.2-25)

plot.sdists.graph: Plotting Edit Transcripts and Sequence Alignments

Description

Function for visualizing the optimal transformation (alignment) graph for two symbol sequences.

Usage

# S3 method for sdists.graph
plot(x, circle.col = 1, graph.col = 2,
     circle.scale = c("mean", "max", "last", "text"), main = "", ...)

Arguments

x

an object of class sdists.graph.

circle.col

color to be used for circles.

graph.col

color to be used for the graph.

circle.scale

scaling to be used for circles.

main

plot title.

...

further unused arguments.

Author

Christian Buchta

Details

This function plots the dynamic programming table, the (back)pointers and the combined graph of optimal edit transcripts (alignments) computed with sdist.trace. The fist sequence is represented by the y-axis and the second by the x-axis.

The circumference of a circle is proportional to the minimum cost (maximum weight) of edit (alignment) operations leading to a table cell. circle.scale specifies the type of normalization performed where last means the last table entry (containing the optimum value), and text draws the values instead of circles.

The (back)pointers, defining possible (traceback) paths, are plotted as dotted edges. Note that a traceback starts in the last cell of the table and ends at the origin.

For the edges of the graph that is the union of all optimal paths, two line types are used: solid for insert, delete, and replace operations, and dashed for a match. The line width indicates the number of times an edge is on a path, but note that the interpretation is device-specific (compare par).

References

D. Gusfield (1997). Algorithms on Strings, Trees, and Sequences. Cambridge University Press, Chapter 11.

Inspired by: http://home.uchicago.edu/~aabbott/.

See Also

sdists.trace, plot

Examples

## continue example in sdists.trace
x1 <- "vintner"
y1 <- "writers"
b11 <- sdists.trace(x1, y1, weight=c(1,1,0,1), graph = TRUE)
b11
plot(b11)
plot(b11, circle.scale = "text")
## partial
b12 <- sdists.trace(x1, y1, weight=c(1,1,0,1), graph = TRUE, partial = TRUE)
b12
plot(b12)