The function tries to turn the dend into hclust. It then uses the cophenetic distance matrix for optimizing the tree's rotation.
This is a good (and fast) starting point for untangle_step_rotate_2side
untangle_DendSer(dend, ...)
An object of class dendlist
NOT USED
A dendlist object with ordered dends
DendSer
, DendSer.dendrogram ,
untangle_DendSer, rotate_DendSer
if (FALSE) { # \dontrun{
set.seed(232)
ss <- sample(1:150, 20)
dend1 <- iris[ss, -5] %>%
dist() %>%
hclust("com") %>%
as.dendrogram()
dend2 <- iris[ss, -5] %>%
dist() %>%
hclust("sin") %>%
as.dendrogram()
dend12 <- dendlist(dend1, dend2)
# bad solutions
dend12 %>% tanglegram()
dend12 %>%
untangle("step2") %>%
tanglegram()
dend12 %>%
untangle_DendSer() %>%
tanglegram()
# but the combination is quite awsome:
dend12 %>%
untangle_DendSer() %>%
untangle("step2") %>%
tanglegram()
} # }