Skip to main content
Figure 6 | BMC Bioinformatics

Figure 6

From: Improving probe set selection for microbial community analysis by leveraging taxonomic information of training sequences

Figure 6

Fidelity effect when removing whole phyla from training data. A) OTU fidelity. B) Genus fidelity. Although both MFPS and MDPS are negatively affected generally by removing whole phyla, the effect is relatively minor, and the MFPS outperforms the MDPS. OTU HFR actually goes up in the MFPS and MDPS when the phyla Proteobacteria and Actinobacteria are removed, respectively. Genus HFR also goes up in the MDPS when Actinobacteria are removed. However, genus HFR decreases in the MFPS when Proteobacteria are removed. The proportions of all training sequences each tested phylum contributes are: Firmicutes (33.5%), Proteobacteria (25.7%), Actinobacteria (24.1%) and Bacteroidetes (9.6%). Error bars are the standard deviations of 100 probe sets.

Back to article page