Table 6.

Notations.

$b$	ASVs preprocessed vector (only the leaves of the cladogram)
$R$	Raw representation matrix
$\hat{R}$	Rearranged representation matrix
$l$	A taxonomic level (Super-kingdom, Phylum, Class, Order, Family, Genus, Species)
$N$	Number of leaves of the cladogram of means
$C_{l,i}$	Hierarchical cluster number $i$ in level $l$
$A$	Adjacency matrix of graph
$\sigma$	Activation function
$W$	Weight matrix in neural network
$I$	Identity matrix
$v$	ASVs frequency vector (all the cladogram’s vertices, and not only the leaves, in contrast with $b$ above)