Tax2Tree-curated NCBI taxonomy
======

The original NCBI taxonomy was curated using Tax2Tree such that the hierarchy
is consistent with the WoL2 phylogeny.

Database files:

 - lineages.txt: Consensus lineage strings with taxon names.
 - linetids.txt: Consensus lineage strings with TaxIDs.
 - decorated.nwk: Phylogeny with nodes decorated with consensus lineages.
 - fmeasures.txt: F-measures of taxa.
 - raw/: raw Tax2Tree output.

Note: This process does not create new taxa (TaxIDs or names). All taxa can be
found in the original NCBI taxonomy. However, some hierarchical relationships
among taxa (species to domain) may be modified in order to be consistent with
the phylogeny.

## Protocol

See ../../README for the Tax2Tree command.

However, pre- and post-processing were needed because the NCBI taxonomy
contains gaps in the lineages (i.e., not all ranks from species to domain
are filled with taxa). For example:

  d__Bacteria; p__Candidatus Tectomicrobia; c__; o__; f__; g__Candidatus
  Entotheonella; s__Candidatus Entotheonella factor

This violates the design of Tax2Tree. To proceed, the empty ranks were filled
with dummy codes based on their parents before running Tax2Tree. These dummy
codes were removed after running Tax2Tree. These operations were performed
using tax2tree_input.py and tax2tree_output.py.