Skip to content

todo.md

Latest commit

 

History

History
91 lines (81 loc) · 2.68 KB

todo.md

File metadata and controls

91 lines (81 loc) · 2.68 KB

To-Do

Repo

  • Organise in branches
  • CI: trigger testing and doc builds selectively

Testing

  • Improve test coverage:
    • Core
    • Newick
    • Bipartitions
    • Distances
  • Use doctests

Documentation

  • Organise docstring pages
  • Add examples (notebooks?)

Functionality

  • tree[branch_id => node_id] syntactic sugar
  • Core functions
    • reroot! (change intrinsic branch orientation)
    • collapse!
    • graft!
    • ungraft
    • rebuild! (remove unlinked elements and re-assign ID numbers)
  • Tree builders
    • Random addition (randtree)
      • Replace method to extend Base.rand instead?
    • Balanced
    • Unbalanced
    • Star
  • Implement neighbour-joining
    • Naive method
    • Dynamic method (Clausen 2022)
  • Bipartitions, distances, dissimilarities &c.
    • Node distance
    • Patristic distance
    • Reduce memory use of BipartitionTable
    • is_refinement method
    • Robinson-Foulds (Penny & Hendy 1985)
      • Basic implementation
      • Normalisations
      • Weighted variants
    • Clustering information distance (Smith 2020)
      • Basic implementation
      • Associated methods (common clustering information, entropy, &c.)
      • Normalisations
      • Reduce LAP allocations
      • Weighted variants?
    • Make distance functions follow the Distances package
      • Implement distance type hierarchy
      • Unify all methods under a dist function (for both tree and node distances)
    • Tree dissimilarities / differences (semi-metric)
      • Contradiction dissimilarity (Bapst et a. 2018)
      • CI contradiction dissimilarity?
  • Implement tree consensus functions
    • Majority-rule (Jensen et al. 2016)
    • Strict (Day 1985?)
  • Implement branch swapping functions
    • Swap nodes
    • NNI
    • SPR
    • TBR
  • Newick writing and reading
    • Write (Newick.write, Newick.string; don't implement print methods)
    • Read (Newick.parse, Newick.read)
    • Read metacomment data
      • Custom readers (RAxML-NG, BEAST2, MrBayes, etc.)
      • Write metacomment data
  • Implement parsimony
    • Data structure for character/sequence data
    • Unweighted Fitch kernel (Goloboff 2022; White & Holland 2011)
    • Weighted Fitch kernel
    • Sankoff kernel
    • Tree scores, ASR
    • Rearrangement scores
    • Search
    • Dynamic kernels

Design

  • Implement taxon deletion from TaxonSet?
  • "Edge-matrix" (APE-style) tree data structure
  • Neighbour-vector (PHYML/GoTree-style) tree data structure
  • Rooted-tree specialisations
    • Time tree