Figure 3

An example of the hierarchical classification process. Four hierarchical levels are constructed to analyze the Arabidopsis thaliana genome (A). Classifiers included are: six alignment-free classifiers at Level 1, 5–10TMs by both Phobius and HMMTOP at Level 2, 7TMs by both Phobius and HMMTOP at Level 3, and SAM2 and GPCRHMM at Level 4. Less stringent filtering can be done by removing the requirement of 7TMs by Phobius and HMMTOP (B). At each level, prediction results can be combined in various ways (C). In the table listing classification results for the A. thaliana genome, each sequence ID is linked to the corresponding gene entry of The Arabidopsis Information Resource (TAIR) website [21, 22] (D).