We are happy to announce the publication of PANORAMA, a new computational tool developed by our team, in PLOS Computational Biology. It has been developed during the PhD thesis of Jérôme Arnoux.
Why PANORAMA?
Over the past decade, the explosion in the number of sequenced bacterial genomes has transformed how we study microbial diversity, evolution, and adaptation. But traditional approaches that analyze genomes one at a time are struggling to keep up with the scale of today’s datasets. The concept of the pangenome — the complete genetic repertoire of a species — offers a powerful alternative, capturing genetic diversity across thousands of strains at once.
PANORAMA builds on this idea. It is designed to exploit pangenome graphs, enabling their annotation and comparison to explore genomic diversity across multiple species.
What does it do?
Built on top of PPanGGOLiN, PANORAMA introduces two main workflows:
- PanSystem — predicts macromolecular systems (such as bacterial defense mechanisms against phages) directly within a pangenome graph, using rule-based models inspired by MacSyFinder. Instead of scanning each genome individually, PANORAMA reasons at the level of gene families, making it possible to analyze thousands of genomes in minutes rather than days.
- PanCompare — compares systems and genomic regions (such as spots of insertion) across the pangenomes of different species, revealing conserved genomic features and shared evolutionary strategies.
Putting it to the test
We validated PANORAMA on a dataset of 941 Pseudomonas aeruginosa genomes, comparing its predictions of phage defense systems against two established reference tools, PADLOC and DefenseFinder. PANORAMA achieved F1-scores of 96–98%, while running dramatically faster — completing analyses in minutes compared to 24–71 hours for the other tools.
Beyond raw performance, PANORAMA automatically rediscovered known “core defense hotspots” in P. aeruginosa and identified two additional ones, by linking predicted systems to regions of genomic plasticity and their insertion spots.
We then scaled the analysis to more than 6,000 genomes across four Enterobacteriaceae species (Escherichia coli, Salmonella enterica, Klebsiella pneumoniae, and Citrobacter freundii). PANORAMA’s comparative workflow uncovered 99 clusters of defense systems conserved across species, offering new insights into how these systems may be shared and exchanged between related bacteria.
Availability
PANORAMA is free, open-source, and designed for easy adoption:
- 📄 Publication: Panorama: A robust pangenome-based method for predicting and comparing biological systems across species, PLOS Computational Biology, 2026.
- 💡Fully reproducible Google Colab notebook: https://github.com/labgem/PANORAMA_article
- 💻 Code: github.com/labgem/PANORAMA
- 📦 Installation: available via bioconda
- 📚 Documentation: panorama.readthedocs.io
We hope PANORAMA will be a valuable resource for microbiologists, bioinformaticians, and evolutionary biologists working to understand the distribution, function, and evolution of biological systems at the pangenome scale. Ready-to-use model collections are also being made available at github.com/PANORAMA-models, and we welcome contributions from the community.