Awesome

CPAN

DBTree - toolkit for megatrees in portable SQL databases

An example mapping of a tree topology to a database table. The mapping is created by processing a Newick tree file (infile.tre) as follows:

megatree-loader -i infile.tre -d outfile.db

With this mapping, several topological queries can be performed quickly when loading the output file in sqlite3 (or the excellent SQLiteBrowser).

-- select the most recent common ancestor of C and F
select MRCA.* from node as MRCA, node as C, node as F 
  where C.name='C' and F.name='F' 
  and MRCA.left < min(C.left,F.left) 
  and MRCA.right > max(C.right,F.right)
  order by MRCA.left desc limit 1;
 
-- select the descendants from node n2
select DESCENDANT.* from node as DESCENDANT, node as MRCA 
  where MRCA.name='n2' 
  and DESCENDANT.left > MRCA.left 
  and DESCENDANT.right < MRCA.right;

Using databases that are indexed in this way, significant performance increases can be accomplished. For example, a very common usage of large, published, static phylogenies is to extract subtrees from them in order to use them for downstream analysis (e.g. in phylogenetic comparative studies). This application is so common that it forms essentially the basis of the success of Phylomatic and the PhyloTastic project. A similar subtree extraction operation is also implemented by NCBI as the option to extract the 'common tree' from the NCBI taxonomy. Here, this functionality is made available by the megatree-pruner program. To benchmark its performance in comparison with a naive approach that operates on Newick strings, a pruner script based on DendroPy was run side by side with the pruner on randomly selected sets of tips from the OpenTree topology. The performance difference is shown below:

Installation

The following installation instructions describe three different ways to install the package. Unless you know what you are doing, the first way is probably the best one. Under any circumstance, it is probably a good idea to familiarize yourself with the capabilities of the cpanm package manager as it offers many more possibilities for controlling the installation (or for fetching, testing, uninstalling, etc.).

1. From the Comprehensive Perl Archive Network (CPAN)

On many Linux-like operating systems as well as MacOSX, the entire installation completes with this single command:

sudo cpanm Bio::Phylo::Forest::DBTree

Advantages - it's simple and all prerequisites are automatically installed. You will obtain the latest stable release from CPAN, which is amply tested.
Disadvantages - you will likely get code that is a lot older than the latest work on this package.

2. From GitHub

On many Linux-like operating systems as well as MacOSX, you can install the latest code from the repository with this single command:

sudo cpanm git://github.com/rvosa/bio-phylo-forest-dbtree.git

Advantages - it's simple, all prerequisites are automatically installed. You will get the latest code, including any new features and bug fixes.
Disadvantages - you will install untested, recent code, which might include new bugs or other features, in your system folders.

3. From an archive snapshot

This is the approach you might take if you want complete control over the installation, and/or if there is a specific archive (such as zenodo release 10.5281/zenodo.1035856) you wish to install or verify.

This approach starts by installing the prerequisites manually:

# do this only if you don't already have these already
sudo cpanm Bio::Phylo
sudo cpanm DBIx::Class
sudo cpanm DBD::SQLite

Then, unpack the archive, move into the top level folder, and issue the build commands:

perl Makefile.PL
make
make test

Finally, you can opt to install the built products (using sudo make install), or keep them in the present location, which would require you to update two environment variables:

# add the script folder inside the archive to the search path for executables
export PATH="$PATH":`pwd`/script
    
# add the lib folder to the search path for perl libraries
export PERL5LIB="$PERL5LIB":`pwd`/lib

BUGS

Please report any bugs or feature requests on the GitHub bug tracker: https://github.com/rvosa/bio-phylo-forest-dbtree/issues

COPYRIGHT & LICENSE

Copyright 2013-2019 Rutger Vos, All Rights Reserved. This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself, i.e. a choice between the following licenses: