beta_diversity_metrics – List of available metrics

Non-phylogenetic beta diversity metrics. These are count based metrics which is based on the OTU table:

• abund_jaccard: abundance weighted Jaccard distance
• binary_dist_chisq: Binary chi-square distance
• binary_dist_chord: Binary chord distance
• binary_dist_euclidean: Binary euclidean distance
• binary_dist_hamming: Binary Hamming distance (binary Manhattan distance)
• binary_dist_jaccard: Binary Jaccard distance (binary Soergel distance)
• binary_dist_lennon: Binary Lennon distance
• binary_dist_ochiai: Binary Ochiai distance
• binary_otu_gain: Binary distance similar to Unifrac G
• binary_dist_pearson: Binary Pearson distance
• binary_dist_sorensen_dice: Binary Sörensen-Dice distance (binary Bray-Curtis distance or binary Whittaker distance)
• dist_bray_curtis: Bray-Curtis distance (normalized Manhattan distance)
• dist_canberra: Canberra distance
• dist_chisq: Chi-square distance
• dist_chord: Chord distance
• dist_euclidean: Euclidean distance
• dist_gower: Gower distance
• dist_hellinger: Hellinger distance
• dist_kulczynski: Kulczynski distance
• dist_manhattan: Manhattan distance
• dist_morisita_horn: Morisita-Horn distance
• dist_pearson: Pearson distance
• dist_soergel: Soergel distance
• dist_spearman_approx: Spearman rank distance
• dist_specprof: Species profile distance

Phylogenetic beta diversity metrics. These metrics are based on UniFrac, which takes into account the evolutionary relationship between sequences:

• dist_unifrac_G: The G metric calculates the fraction branch length in the sample i + sample j tree that is exclusive to sample i and it is asymmetric.
• dist_unifrac_G_full_tree: The full_tree version calculates the fraction of branch length in the full tree that is exclusive to sample i and it is asymmetric.
• dist_unweighted_unifrac: This is the standard unweighted UniFrac, which is used to assess ‘who’s there’ without taking in account the relative abundance of identical sequences.
• dist_unweighted_unifrac_full_tree: Typically, when computing the dissimilarity between two samples, unifrac considers only the parts of the phylogenetic tree contained by otus in either sample. The full_tree modification considers the entire supplied tree.
• dist_weighted_normalized_unifrac: Weighted UniFrac with normalized values and is used to include abundance information. The normalization adjusts for varying root-to-tip distances.
• dist_weighted_unifrac: Weighted UniFrac.