Skip to main content
Figure 3 | BMC Bioinformatics

Figure 3

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

Figure 3

Optimizing the genus penalty for the MFPS cost function. The genus penalty was varied from 0 to 200 while holding the OTU penalty at 1. A) OTU fidelity rises slightly as the genus penalty increases from zero and then declines. B) Genus fidelity rises sharply and then plateaus. A penalty of 10 achieves the highest OTU fidelity in the 20% cross-validation data (CV) and OTU fidelity is highest in the 100% data set at a penalty of 30. Error bars are the standard deviations of 100 probe sets per data point. Note: for display purposes, the y-axis scales are not identical in A and B.

Back to article page