For example, analysing sequences derived from members of the
Verrucomicrobia, results in the plot shown in Fig. 1. Each
data point represents a separate pairwise comparison.
In this example, 97 sequences were analysed, resulting in 4,656
separate comparisons.
Most of the data points in Fig. 1 cluster relatively
close to the x-axis, and reflect comparisons between error-free
sequences. However, there are also some unusually high values
indicating that some comparisons involve anomalous
sequences
. To clarify which data points should be treated as outliers, a suitable cut-off line is superimposed
over the data (Fig. 2). Datapoints above the cut-off line are
identified as
outliers. Sequences responsible for these outliers can now be identified.
Figure 2. A cut-off line
is superimposed over the data points to identify which values should be
treated as outliers. Data points
below the cut-off line are judged to
result from comparisons between error-free sequences. Data points
above the line signify anomalous sequences.
Note...
Right mouse-clicking the graph image calls
up a
pop-up menu that allows the user to view the raw data, print the plot
or save the plot image.
