Dotplots
The concept of similarity can be illustrated by a simple dot plot. Dot plots
are used to visually compare two sequences and detect of regions of close
similarity between them.
-
In Figure 1 below, two sequences are arranged along the axes of a simple graph.
At every point where the two sequences are identical, a dot is placed (i.e. at the intersection of every row and column that have the same letter in both
sequences).
Figure 1. Simple dot plot
-
A diagonal stretch of dots will indicate regions where the two sequences are
similar. A trained eye can pick out a diagonal pattern of similarity in Figure 1, but statistical methods can be applied to make the results more apparent.
A filter can be used to place a dot only when a group of successive bases match.
-
In Figure 2 the same dot plot is shown with a filter such that a dot is printed
only if in a window of 4 bases, 3 of these 4 bases match.
-
To detect more distant similarities, it may be better to use a
much larger window (ie. 20, 30, or even 50 bases) and some suitable percentage of identities (perhaps 50%).
Figure 2. Dot plot with a 4 base 75% identity window
However, for real data, these patterns may not be quite so obvious.
Figure 3. A sample dot plot of two 230 AA database sequences
