Skip to main content
. 2020 May 29;36(16):4383–4388. doi: 10.1093/bioinformatics/btaa548

Fig. 1.

Fig. 1.

Illustration of the taxon-specific, RAR-based GO term misannotation detection approach. The area of the rectangle is proportional to the number of proteins annotated with a GO term (row) in a taxon (column). The first two GO terms are unlikely to be inconsistently annotated because different taxa have similar the portion of annotated proteins. The next two GO terms are also disregarded by our analysis because each GO term is only annotated to one taxon. The last two GO terms will be picked up by the RAR analysis because it is common in at least one taxon but rare in another taxon