Bridging Gene-level And Sequence-level Pangenome Graph To Explore Microbial Diversity
Résumé
Pangenomics aims to study species diversity by collecting and comparing all genomic sequences of a species. Initially, pangenomic studies focused on microbial diversity by defining sets of gene families. However, pangenome representation has since evolved towards a graph-based approach to better capture genomic diversity. These graphs can be analyzed at different levels, either at the gene level or at the sequence level (e.g., compacted de Bruijn graphs or variation graphs). The aim of the present work is to augment gene-level pangenome graphs by adding a new layer of resolution down to the single-nucleotide level, while remaining compatible with the PPanGGOLiN method. While PPanGGOLiN provides a global perspective of the pangenome at the gene family level, we extend its capabilities to incorporate both coding and non-coding genes, as well as the panIGRome (panInterGenic Region), to enable a full-genome representation of the input genomes. We then build a compacted de Bruijn graph for each gene family and PPanGGOLiN edge, and connect all these local subgraphs for in-depth visualization through the Vizitig visualization method. Vizitig enables graph manipulation and exploration, including subgraph selection via relational queries, graph coloring, and graph export, thereby facilitating navigation over this two-level pangenome graph. This development unifies gene-level and sequence-level representations, offering deeper insights into genome diversity. Our next steps involve mapping long-read sequencing data onto the integrated graph to enable simultaneous detection of variants and gene presence.
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| licence |
