Interpretable metric learning in comparative metagenomics: The adaptive Haar-like distance | PLOS Computational Biology
Skip to main content
Advertisement

< Back to Article

Interpretable metric learning in comparative metagenomics: The adaptive Haar-like distance

Fig 8

Comparison of the adaptive Haar-like embedding to various phylogenetic β-diversity metrics in the Autism dataset.

A: Haar-like Distance PCoA embedding (F = 6.66). B: Unweighted UniFrac PCoA embedding (F = 18.44). C: Weighted UniFrac PCoA embedding (F = 7.80). D: Adaptive Haar-like PCoA embedding using 2 Haar-like coordinates (F = 34.96). E: Box plots of the top two Haar-like coordinates across the various diet types. F: The three most important Haar-like coordinates of the Autism dataset visualized on Greengenes 97%.

Fig 8

doi: https://doi.org/10.1371/journal.pcbi.1011543.g008