PhyloDistance-CI: Clustering Information Distance

In npcooley/SynExtend: Tools for Working With Synteny Objects

Clustering Information Distance

Description

Calculate distance between two unrooted phylogenies using mutual clustering information of branch partitions.

Details

This function is called as part of PhyloDistance and calculates tree distance using the clustering information approach first described in Smith (2020). This function iteratively pairs internal tree branches of a phylogeny based on their similarity, then scores overall similarity as the sum of these measures. The similarity score is then converted to a distance by normalizing by the average entropy of the two trees. This metric has been demonstrated to outperform numerous other metrics in capabilities; see the original publication cited in References for more information.

Users may wish to use the actual similarity values rather than a distance metric; the option to specify RawScore=TRUE is provided for this case. Distance is calculated as \frac{M - S}{M}, where M=\frac{1}{2}(H_1 + H_2), H_i is the entropy of the i'th tree, and S is the similarity score between them. As shown in the original publication, this satisfies the necessary requirements to be considered a distance metric. Setting RawScore=TRUE will instead return a vector with (S, H_1, H_2, p), where p is an approximation for the two sided p-value of the result based on random simulations from Smith (2020).

Value

Returns a normalized distance, with 0 indicating identical trees and 1 indicating maximal difference. Note that branch lengths are not considered, so two trees with different branch lengths may return a distance of 0.

If RawScore=TRUE, returns a named length 4 vector with the first entry the similarity score, subsequent entries the entropy values for each tree, and the last entry the approximate p-value for the result based on simulations.

If the trees have no leaves in common, the function will return 1 if RawScore=FALSE, and c(0, NA, NA, NA) if TRUE.

Note

Note that this function requires the input dendrograms to be labeled alike (ex. leaf labeled abc in dend1 represents the same species as leaf labeled abc in dend2). Labels can easily be modified using dendrapply.

Author(s)

Aidan Lakshman [e-mail]

References

Smith, Martin R. Information theoretic generalized Robinson–Foulds metrics for comparing phylogenetic trees. Bioinformatics, 2020. 36(20):5007-5013.

Examples

# making some toy dendrograms
set.seed(123)
dm1 <- as.dist(matrix(runif(64, 0.5, 5), ncol=8))
dm2 <- as.dist(matrix(runif(64, 0.5, 5), ncol=8))

tree1 <- as.dendrogram(hclust(dm1))
tree2 <- as.dendrogram(hclust(dm2))

# get RF distance
PhyloDistance(tree1, tree2, Method="CI")

# get similarity score with individual entropies
PhyloDistance(tree1, tree2, Method="CI", RawScore=TRUE)

npcooley/SynExtend documentation built on Dec. 20, 2024, 4:03 p.m.