➔ Slides
Authors
Abstract
Phylogenetic analyses aim at reconstructing the evolutionary history of biological objects from molecules to species, and populations. Faced with the number of programs available and the difficulty for scientists to combine them, we designed in 2008 Phylogeny.fr, which has quickly become one of the most used platforms to perform phylogenetic analyses. However, due to the diversity of analyses performed (phylogeny.fr can be simultaneously used by hundreds of students or can be used through batch scripts), the number of analyses performed (50,000 per month), and the number of new phylogenetic tools available, the need to refactor Phylogeny.fr has become crucial.
In this talk, we introduce NGPhylogeny.fr (Next Generation Phylogeny.fr), developed within a Python Web framework (Django), in which we have refactored Phylogeny.fr and made it distributable by designing a scalable environment, an easy-to-use web interface based on a series of modular Galaxy workflows able to perform a very large variety of phylogenetic analyses. Moreover, we have performed a reproducibility study, to systematically compare the results obtained by the Galaxy-based NGPhylogeny.fr workflow and the original phylogeny.fr, using real datasets.
Our talk will highlight how (i) NGPhylogeny.fr can be used in a functional genomics context to quickly analyze large sets of protein superfamilies, (ii) in-depth studies can be quickly launched and (iii) NGPhylogeny.fr can be installed on a wide variety of configurations. On a more generic aspect, we will underline the benefit of designing a coupled Django-interface / workflow-Galaxy environment for end-users.