In ms609/TreeDistData: Analysis of Phylogenetic Tree Distance Measures

Method

Take an unrooted ten-leaf tree: ```{R backbone, echo=FALSE, fig.width=2, fig.asp=1.2, message=FALSE, fig.align='center'} library('Quartet', exclude = 'RobinsonFoulds') library('TreeTools', quietly = TRUE, warn.conflicts = FALSE) library('TreeDist') library('TreeDistData')

origPar <- par(mar = rep(0, 4))

backbone <- ape::read.tree(text='(a, (b, (c, (d, (((e, f), g), (h, (i, j)))))));') plot(UnrootTree(backbone)) nTip <- length(backbone$tip.label) add3 <- structure(lapply(lapply(AddTipEverywhere(backbone, '11th leaf'), AddTip, 11L, '12th leaf'), AddTip, 12L, '13th leaf'), class = 'multiPhylo') add1then2 <- lapply(AddTipEverywhere(backbone, '11th leaf'), function (tr) structure(lapply(AddTipEverywhere(tr, '12th leaf'), AddTip, 12L, '13th leaf'), class='multiPhylo')) add2then1 <- lapply(AddTipEverywhere(backbone, '12th leaf'), function (tr) structure(lapply(AddTipEverywhere(tr, '11th leaf'), AddTip, 11L, '13th leaf'), class = 'multiPhylo') )

The backbone tree has `r length(backbone$tip.label)` leaves;
`r length(add3)` test trees were generated by adding an clade of three 
leaves at each of the `r length(add3)` edges of the unrooted backbone tree.

These are trees T<sub>1</sub>--T<sub>17</sub>.

```r
par(mfrow = c(3, 6), mar = rep(0.2, 4))
urt <- lapply(add3, unroot)
lapply(urt, plot) -> XX

For each of these trees -- we'll consider T₁ first -- we can then move the eleventh leaf to each of the other edges of the tree, giving trees U_1.1-U_1.19:

par(mfrow = c(4, 5), mar = rep(0.2, 4))
urt <- lapply(add1then2[[1]], unroot)
lapply(urt, plot) -> XX

or the cherry containing the twelfth and thirteenth, giving trees V_1.1 to V_1.19:

par(mfrow = c(4, 5), mar = rep(0.2, 4))
urt <- lapply(add2then1[[1]], unroot)
lapply(urt, plot) -> XX
par(origPar)

Two trees occur in both U_i and V_i, i.e. they can be generated by moving one leaf or moving two leaves. These trees are excluded from further analysis.

We then expect the distance from T_i to U_i.j (i.e. move one tip to a new location) to be less than the distance from T_i to V_i.j (i.e. move two tips to the same location)

baddies <- vapply(seq_along(add3), function(i) {
  move1 <- AllDists(add2then1[[i]], add3[[i]])
  move2 <- AllDists(add1then2[[i]], add3[[i]])
  move1[c('mast', 'masti'), ] <- -move1[c('mast', 'masti'), ]
  move2[c('mast', 'masti'), ] <- -move2[c('mast', 'masti'), ]
  rowSums(move2 <= move1)
}, double(25))

Results

errors <- t(t(rowSums(baddies) - (2L * length(add3))))
dimnames(errors) <- list(tdMdAbbrevs[rownames(errors)], 'Inconsistencies')
.TDDTable(DT::datatable, errors)

ms609/TreeDistData documentation built on June 30, 2024, 7:21 p.m.