fastsubtrees 1.6

Tree representation for fast queries of the list of IDs of any subtree

Fastsubtrees

Fastsubtrees is a Python library and a command line script, for handling fairly large trees (in the order of magnitude of millions nodes), in particular allowing the fast extraction of any subtree.

The main application domain of fastsubtrees is working with the NCBI taxonomy tree, however the code is implemented in a generic way, so that other applications are possible.

For achieving an efficient running time and memory use, the nodes of the tree are represented compactly in deep-first traversal order. Subtrees are then extracted in O(s) time, where s is the size of the extracted subtree (i.e. not depending on the size of the whole tree).

The tree representation can be saved to file, so that it must be not be re-computed each time. It is dynamical, i.e. after a tree has been created, it can be modified, by adding a new leaf node, or an entire subtree under an existing node. Also, existing leaf nodes or entire subtrees can be deleted.

Along with numerical node identifiers, additional information can be stored for each node, by defining any number of optional node attributes.

Node identifiers

Each node must be represented by a unique positive ID. The IDs must not necessarily be all consecutive (i.e. some "holes" may be present), but the largest node ID (idmax) should not be much larger than the total number of nodes, because the memory consumption is in O(idmax).

The NCBI taxonomy tree, for example, fullfills the conditions stated above.

However, the library can be used on any other tree. For this, if the IDs are non-numerical, contain zero or negative numbers, or the ID space is not compact, they must be first mapped to different IDs.

Attributes

Besides a numerical node identifier, for each node of the tree additional information can be stored in form of attributes. Any number of attributes can be created. Each node can contain a single value or a set of values for an attribute, but there is no requirement that all nodes have values for an attribute.

For each attribute defined in a tree, a file is created, where the attribute values are stored. The attributes are also stored in deep-first traversal order, so that the list of attribute values for an entire subtree can be queried efficiently.

Tree construction

For the construction of the tree a data source for the tree node identifiers must be provided. For each node, the data source must provide the ID of the node itself and of its parent. No particular order of the data is required, i.e. a child node information may be provided before or after its parent node. The data source can be e.g. a database table or a tabular file (or anything else).

The same interface is used when adding a new subtree (with the difference, in the current implementation, that in this case child nodes must be provided after their parent node).

Working with the library

Installation

The package can be installed using pip install fastsubtrees.

Docker

To try or test the package, it is possible to use fastsubtrees by employing the Docker image defined in Dockerfile. This does not require any external database installation and configuration.

# create a Docker image
docker build --tag "fastsubtrees" .

# create a container and run it
docker run -p 8050:8050 --detach --name fastsubtreesC fastsubtrees
# or, if it was already created and stopped, restart it using:
# docker start fastsubtreesC

# run the tests
docker exec fastsubtreesC tests

# run the benchmarks, skipping repeating tree creation
docker exec fastsubtreesC benchmarks

# run all benchmarks, including tree creation
docker exec fastsubtreesC benchmarks --all

# run the example application
docker exec fastsubtreesC start-example-app
# now open it in the browser at https://0.0.0.0:8050

Tests

To run the test suite, you can use pytest (or make tests). The tests include tests of fastsubtrees and of the sub-package ntmirror. The latter are partly dependent on a database installation and configuration which must be given in ntmirror/tests/config.yaml; database-dependent tests are skipped if this configuration file is not provided.

The entire test suite can be also run from the Docker container, without further configuration, see above the Docker section.

Benchmarks

Benchmarks can be run using the shell scripts provided under benchmarks. These require data, which is downloaded from NCBI taxonomy and some pre-computed example data which is provided in the data subdirectory (genome sizes and GC content).

The benchmarks can be convienently run from the Docker container, without requiring a database installation and setup, see above the Docker section.

The _all version also benchmarks the construction of the representation of the NCBI taxonomy tree and requires about 40-70 minutes to complete, depending on the system.

Command line interface

The command line tool fastsubtrees allows constructing a tree, modifying it (by adding and deleting leaf nodes or subtrees), adding attributes, and performing a subtree query (listing IDs or attribute values in a subtree).

The scripts are designed to be very flexible: e.g. the data source for tree construction, or for obtaining attribute values, can be freely defined by the user and passed to the scripts as Python code and configuration data. Modules for the most common cases (database, tabular file) are provided.

The command line interface is further described in the CLI manual.

CLI example: NCBI taxonomy tree

This is an example of the basic operations using the NCBI taxonomy tree:

# download the NCBI taxonomy database dumps
ntmirror-download ntdumps
# construct the fastsubtrees tree data structure
fastsubtrees construct nt.tree --ntdump ntdumps
# query the IDs under node 562
faststubrees query nt.tree 562

The following adds attributes from the example data (GC content and genome size of Bacterial genomes) stored in the repository:

# add a genome size attribute from a tabular file
# the IDs are in column 1, the values in column 2 of the table
fastsubtrees attr construct nt.tree genome_size --tab \
  data/accession_taxid_attribute.tsv.gz 1 2

# add a GC content attribute from a tabular file
# the IDs are in column 1, the values in column 3 of the table
fastsubtrees attr construct nt.tree GC_content --tab \
  data/accession_taxid_attribute.tsv.gz 1 3

Once the attributes are created, their values in any subtree can be easily queried:

# query the genome size values under node 562
fastsubtrees attr query --ids nt.tree genome_size 562

CLI example with generic data

Fastsubtrees is not only usable with the NCBI taxonomy tree. The following example constructs a tree with example data included with the repository loading it from a tabular file.

# construct the tree, using a tabular file as data source
fastsubtrees construct small.tree --tab tests/testdata/small_tree.tsv

The data source can be differet: for example also a differently formatted tabular file or a database table. For these cases, a Python module is passed, which yields the tree data, as described in the CLI manual.

API

The library functionality can be also directly accessed in Python code using the API, which is documented in the API manual.

Subpackages

ntmirror

When working with the NCBI taxonomy database, a local copy of the NCBI taxonomy dump can be obtained and kept up-to-date using the ntmirror package, which is located in the directory ntmirror. It is a Python package, which can be installed using pip, separately from fastsubtrees.

Besides downloading the dump, when needed, the package also allows to load the NCBI taxonomy database in a local SQL database, and to extract subtrees from it using hierarchical SQL queries.

Please refer to the user manual of ntmirror located under ntmirror/docs for more information.

Genomes Attributes Viewer

An interactive web application based on fastsubtrees was developed using dash. It allows to graphically display the distribution of values of attributes in subtrees of the NCBI taxonomic tree.

This example application is located under genomes-attributes-viewer. For more information see the genomes-attributes-viewer/README.md file.

The application can be conveniently setup and started using the Docker image, see above in the Docker section.