Skip to the content.

Tiara

Deep-learning-based approach for identification of eukaryotic sequences in the metagenomic data powered by PyTorch.

The sequences are classified in two stages:

For more information, please refer to our paper: Tiara: Deep learning-based classification system for eukaryotic sequences.

Installation

Requirements

Quick installation

Detailed instructions can be found here.

Using pip

Run pip install tiara, preferably in a fresh environment.

Using conda

Run conda install -c conda-forge tiara, preferably in a fresh environment.

We recommend to use mamba instead of conda (it’s faster).

Unfortunately currently it does work only for python 3.7 and 3.8.

Using setup.py

Latest stable release
Latest developer version
git clone https://github.com/ibe-uw/tiara.git
cd tiara
python setup.py install

Usage

Sample usage can be found here.

Sample pipelines

Here we describe some pipelines to tackle metagenomic data that utilize tiara.

Citing Tiara

Michał Karlicki, Stanisław Antonowicz, Anna Karnkowska, Tiara: deep learning-based classification system for eukaryotic sequences, Bioinformatics, Volume 38, Issue 2, 15 January 2022, Pages 344–350, https://doi.org/10.1093/bioinformatics/btab672

License

Tiara is released under an open-source MIT license

Name

In the Polish translation of the Harry Potter book series, the Sorting Hat (which assigned wizards to different houses) was called Tiara Przydziału. We thought that it’s an appropriate name for a software which classifies sequences to different taxonomic units. In English the word tiara usually refers to a papal tiara. A papal tiara has three crowns, and life has three domains, so maybe that’s another explanation for the name of our program.

Version history: