Figure 5.

The p-value distributions of sequence pairs annotated for cellular component. Each curve represents the percentages of sequence pairs of less than or equal to certain p-value across different levels of the GO category. (Some curves do not include the percentages for all levels because no sequence pair on those levels has a p-value less than or equal to certain thresholds.)

Duan et al. BMC Bioinformatics 2006 7(Suppl 4):S11   doi:10.1186/1471-2105-7-S4-S11