Skip to main content
. 2008 May 30;9:255. doi: 10.1186/1471-2105-9-255

Figure 3.

Figure 3

Scatter plot for the sequence homology-based method, showing percent correct versus sequence identity for the 742 pairs of testing sequences and their corresponding homologs. A 5-fold cross-validation was performed with 5 samples of 200 sequences each, and the combined results for all of the 742 sequences with homologs are plotted here. Sequence identity was computed as the number of identical residues divided by the query sequence length.