Automated reconstruction of all gene histories in large bacterial pangenome datasets and search for co-evolved gene modules with Pantagruel

The availability of bacterial pangenome data grows exponentially, requiring efficient new methods of analysis. Currently popular approaches for the fast comparison of genomes have the drawback of not being based on explicit evolutionary models of diversification. Making sense of bacterial genome evolution, and notably in the accessory genome, requires however to take into account the complex processes by which the genomes evolve. Here we present the Pantagruel bioinformatic software pipeline, which enables the construction of a complete bacterial pangenome database geared towards the inference of gene evolution scenarios using gene tree/species tree reconciliation. Pantagruel is a modular pipeline that combines state-of-the-art external software with unique new methods. It can be executed with no supervision to perform a standard pangenome analysis, or be configured by advanced users to integrate methods of choice. A relational database underlies its data structure, allowing efficient retrieval of the large-scale data generated by integrative analyses of pangenome evolutionary history. From the reconstructed gene evolution scenarios, two main outputs are derived: firstly the gene tree-aware assignation of orthology, allowing the fine analysis of gene gain and loss history over the species phylogeny, and secondly a network of gene-to-gene association based on correlated events in scenarios of gene evolution, leading to the definition of co-evolved gene modules. Pantagruel is available as an open source software package at https://github.com/flass/pantagruel.