ms609 / treesearch Goto Github PK

Introduce:

• Installing R
• Installing packages

#' @template InitializeDataParam
#' @template ClearUpDataParam
#' @template TreeScorerParam

A substantial proportion of time is being spent sampling, and in the overhead of decomposing a tree into variables. If we could list all trees that were 1 SPR step away from the current tree and score those, we'd save a lot of time in rearrangements.

Should be straightforward, and will reduce duplication

Reproduce by:

• Run EasyTrees()
• Map character
• Set a root
• Tips aligned regardless of "Align tips" value

And update inst/CITATION

Hi,

I update the package to version 0.4.0 . But it doesn't work. Below is the error

library(TreeSearch)
Warning message:
package ‘TreeSearch’ was built under R version 3.5.3
SampleOne(1:10,len=10)
Error in SampleOne(1:10, len = 10) : could not find function "SampleOne"

Do this by first matching all identical splits, then only calculating the information content of non-identical split pairings.

Use TreeTools function; see TreeTools#53.

https://cran.r-project.org/web/packages/Rdpack/vignettes/Inserting_bibtex_references.pdf

See "decisiveness" branch.

• app.R
• inst/References.bib

IW now active; integrate into vignette

Should cite TreeSearch as standalone package, with MorphyLib & algorithm paper to be cited when used (cf. profile parsimony / successive approx / etc context)

Was other colour customization planned too?

Adapting the functions developed for Inapp.
Recall that hcl.colors() is available for nice visualizations.

Test case is disabled.

• IPC
• Promising approaches
• Alternative approach (incompatible with progress bars?)

• Support stats like:

  • Bremer/branch supports
  • CI/RI: See phangorn:CI()
  • bootstrapping: See MorphyBootstrap(), though Jackknife() is usually more appropriate.

• Ability to

  • 'force' a tip into a clade
  • & run a Templeton test for alternative hypotheses
  • [MS note for v1.0:] Can profile parsimony offer a better test?

Concordance.R should be completed or omitted before release of v1.0.0

See Minh et al. (2020)

Requires scores to be calculated on polytomies.

See Kearney 2002, Syst Biol, for a potential use wrt rogue taxa.

@article{Nixon1996,
  title = {On consensus, collapsibility, and clade concordance},
  author = {Nixon, Kevin C. and Carpenter, James M.},
  year = {1996},
  journal = {Cladistics},
  volume = {12},
  number = {4},
  pages = {305--321},
  doi = {10.1111/j.1096-0031.1996.tb00017.x}
}

@article{Kearney2002,
  title = {Fragmentary taxa, missing data, and ambiguity: mistaken assumptions and conclusions},
  author = {Kearney, Maureen},
  year = {2002},
  journal = {Systematic Biology},
  volume = {51},
  number = {2},
  pages = {369--381},
  doi = {10.1080/10635150252899824}
}

Once in play, make following updates:

• StepInformation() – provide alternative to Carter1()
• ConcordantInformation(): update docs

Review and implement appropriate methods reviewed in Grant & Kluge (2003)

In Rd documentation for the functions themselves, as well as in vignette

Decide this, then add to DESCRIPTION

Can we implement a third? A fourth?

Hi Martin,

My name is Eunok Lee from the Westmead Institute for Medical Research. I am trying to load the tree using ape and phytool packages and I ran into the issues below:

1. When I tried to load my phylogenetic tree using read.tree(mytree.newick), I found that some tip labels were removed. When I opened the same newick file in text, I could see the tip label that I am interested in. I reinstalled the ape package and re-loaded in R, but the same problem occurred.... Could you help me how to fix this issue?

2. Due to the problem above, I tried using phytools::read.newick(). This does import the tree with all the tip labels; however, the tip label is only showing up to 22 characters. What should do to make the tip label to display the tip labels in a full character length?

Thank you and I look forward to your advice!

With kind regards,

Eunok

Possibly implemented in phangorn; add to app.R

It, and related functions, should at least display a message when DNA data is fed to it.

Save converting from edge to and from morphy's parentof/left/right notation by performing tree reordering using this notation.

hopefully we don't need it any more

the old mpl_visualise functions might be given new life using our Inapp visualizer.

Check that all nodes are bifurcating before running NNI/TBR/SPR, which will crash R.

Tree: hy_iw_k4_16.868_ERROR.txt

Dataset: X24932.txt

Warning message:
In if (homoplasies < 0) stop("Minimum steps have been miscalculated.\n Please report this bug at\n https://github.com/ms609/TreeSearch/issues/new") :

Either

• Remove rgl from imports, and from the app
  or
• Add 3d plotting functionality to the app

Check whether using order(method='radix') improves search speed (https://csgillespie.github.io/efficientR/efficient-performance.html)

in phylo.R:
name[order(tips)] <- tipLabel[tips]
edge2[isTip] <- order(tips)

Function doesn't work with
dataset <- inapplicable.datasets[['Longrich2010']]

Need to support it.

See branch: pp-vignettes

Turbinicarpus ss_26 paupout.txt

I have tested the package with some data (attached).
Here the code that causes the crash. This snippet is taken from vignettes/inapplicable.rmd example code, which runs flawlessly.

turbini.data <- ReadAsPhyDat("Turbinicarpus ss_26 paupout.nex")
RNGversion("3.5.0")
set.seed(0)

nj.tree <- NJTree(turbini.data)
outgroup <- c("beguinii")
rooted.tree <- EnforceOutgroup(nj.tree, outgroup)

better.tree <- TreeSearch(tree=rooted.tree, dataset=turbini.data,
                        EdgeSwapper=RootedNNISwap, verbosity=3)

Thank you very much for the help.

Function to complement [MinimumLength()] is advertised in NEWS.md but doesn't yet exist.

Possibly venues: JOSS; R journal