Insights into goatpox virus and sheeppox virus genomes from pangenome graphs

This study utilizes pangenome variation graphs alongside phylogenetic and gene-level analyses to reveal distinct evolutionary histories between goatpox and sheeppox viruses, highlighting how structural variation at inverted terminal repeats drives genomic diversity and host specificity in these economically important pathogens.

The Gist

The Big Picture: A Tale of Two Viral Cousins

Imagine the world of animal viruses as a neighborhood. In this neighborhood, there are three cousins who cause trouble for livestock: Goatpox (GTPV), Sheeppox (SPPV), and Lumpy Skin Disease (LSDV). They are all "poxviruses," meaning they are large, complex DNA viruses that look very similar on the outside but have some crucial differences in how they behave and who they infect.

This study is like a detective story where scientists decided to stop looking at these viruses one by one and instead looked at their entire family history and genetic blueprints all at once. They used a new, high-tech tool called a "Pangenome Variation Graph" (PVG).

The Analogy: The Family Tree vs. The Subway Map

• Old Way (Phylogenetics): Traditionally, scientists drew a "family tree" to show how viruses are related. This is like a simple tree diagram showing who is the parent of whom. It's good, but it's flat. It assumes everyone follows a single straight line of descent.
• New Way (Pangenome Graphs): This study used a "PVG," which is more like a complex subway map. Instead of one single line, imagine a map where many different train routes (genomes) run through the same city. Some routes share tracks (common genes), but some have unique detours, loops, or dead ends (mutations and structural changes). This map shows not just who is related, but where the paths diverge and how the tracks are built.

The Main Discovery: Two Different Personalities

The researchers found that Goatpox and Sheeppox have very different "personalities" when it comes to their evolution.

1. Goatpox (GTPV): The Old, Stable Family

• The Vibe: Think of Goatpox as an old, established family with three distinct branches that have been living in separate houses for a very long time.
• The Finding: The virus is divided into three deep, stable groups (clades). They rarely mix with each other. It's like three cousins who haven't spoken in decades; their genetic "houses" are very different.
• The "Closed" Book: The study found that Goatpox has a "closed pangenome." Imagine a library where you've read almost every book. If you add one more book to the collection, it's unlikely to be something totally new; it's probably just a copy of something you already have. This means scientists have a pretty good handle on Goatpox; finding new samples won't reveal many surprises.

2. Sheeppox (SPPV): The Young, Chaotic Crowd

• The Vibe: Sheeppox is like a younger, more chaotic family that recently had a massive population boom.
• The Finding: The groups within Sheeppox are messy and overlapping. They mix and mingle more. It's like a crowded party where everyone is talking to everyone else.
• The "Open" Book: Sheeppox has an "open pangenome." Imagine a library that is still being built. Every time you add a new book (a new virus sample), it's likely to be a brand new genre or a unique story you've never seen before. This suggests there is still a lot we don't know about Sheeppox, especially because we haven't looked at enough samples from Africa yet.

The "Fringe" of the Genome: The Tail and the Head

Viruses have a "core" (the middle part) that is essential for survival, and "ends" (the tails) that are more flexible.

• The Core: This is the engine room. It's the same in almost every virus, like the chassis of a car.
• The Ends (ITRs): The ends of the virus genome are like the decorative bumper and spoiler on a car. They aren't essential for the car to drive, but they determine how the car looks, how fast it goes, and who it can talk to.
• The Discovery: The study found that the most wild, crazy changes happen at these "ends."
  • In Goatpox, the ends are different between the three families, acting like a genetic fence keeping them apart.
  • In Sheeppox, the ends are constantly rearranging themselves. Sometimes genes get cut short (truncated), sometimes they get stretched out, and sometimes they merge with neighbors. This "structural plasticity" is likely how the virus adapts to new hosts or evades vaccines.

Why Does This Matter?

1. Vaccines and Protection
We know that a vaccine for Goatpox often protects against Sheeppox (and vice versa), but the Lumpy Skin Disease vaccine is a bit trickier. Understanding these genetic "end" differences helps scientists figure out why some vaccines work better than others and how to design better ones.

2. The "Missing" African Samples
The study highlighted a gap in our knowledge. We have lots of samples from Europe and Asia, but very few from Africa. Since Sheeppox seems to be "open" (always changing), the African samples might hold the keys to understanding how the virus spreads and evolves. It's like trying to solve a puzzle with half the pieces missing.

3. Better Tools for the Future
By building these "Subway Maps" (PVGs), the researchers created a better reference guide for future scientists. Instead of trying to force a new virus sample to fit into an old, rigid template, they can now see exactly where it fits on the map. This makes it easier to spot dangerous mutations quickly.

The Takeaway

This paper tells us that while Goatpox and Sheeppox look similar, they are evolving in very different ways. Goatpox is a stable, ancient family with clear boundaries, while Sheeppox is a dynamic, rapidly changing crowd that is still figuring out its identity.

By using a new "subway map" approach to look at their DNA, scientists can see the hidden structural changes that traditional methods miss. This is crucial for keeping our livestock safe, designing better vaccines, and understanding how these viruses jump between animals.

Technical Summary

1. Problem Statement

Capripoxviruses (CaPV), comprising Goatpox virus (GTPV), Sheeppox virus (SPPV), and Lumpy Skin Disease Virus (LSDV), are economically devastating, transboundary double-stranded DNA (dsDNA) pathogens affecting livestock. While previous studies have characterized the core genomes and broad population structures of these viruses, significant gaps remain regarding:

• The mechanisms driving genomic diversification and host specificity between GTPV and SPPV.
• The contribution of non-coding regions and structural variation (particularly at the Inverted Terminal Repeats or ITRs) to viral evolution.
• The limitations of linear reference-based analyses in capturing complex variation in large DNA viruses.

2. Methodology

The study employed an integrative framework combining phylogenetics, population genetics, and Pangenome Variation Graphs (PVGs) to analyze high-quality genome assemblies.

• Data Collection:
  • GTPV: 14 complete genomes.
  • SPPV: 30 complete genomes (excluding two low-quality assemblies).
  • LSDV: Used as an outgroup and reference (121 genomes for comparative selection analysis) to standardize gene nomenclature (based on the Oman LSDV reference, PV877838).
• Phylogenetic & Population Analysis:
  • Alignments performed using MAFFT.
  • Phylogenetic trees reconstructed using RAxML-NG (GTR+G model).
  • Principal Component Analysis (PCA) and network-based community detection to assess population structure.
• Pangenome Variation Graph (PVG) Construction:
  • Tools: Panalyze, PGGB (v0.4.0), wfmash, and ODGI.
  • Graphs constructed in GFA format to represent SNPs, indels, and structural variations simultaneously.
  • Pangenome Openness: Estimated using Panacus and pangrowth (k-mer and sequence-based methods) to calculate the $\alpha$ parameter (Heaps' Law) determining if the pangenome is "open" (new mutations found with more sampling) or "closed."
• Selection Analysis:
  • Calculated the Direction of Selection (DoS) metric ($DoS = \frac{D_N}{D_N+D_S} - \frac{P_N}{P_N+P_S}$) to distinguish between neutral evolution, purifying selection, and adaptive evolution.
  • Compared ancestral substitutions ($D_N, D_S$) against population-level polymorphisms ($P_N, P_S$).
• Annotation & Visualization:
  • Genomes annotated with Prokka; PVGs visualized using Bandage and SequenceTubeMap.
  • Specific focus on ITR regions (5' and 3' ends) and genes homologous to LSDV ORFs 1–23 and 124–156.

3. Key Contributions

• Graph-Based Genomics for Poxviruses: Demonstrated the utility of PVGs in resolving complex structural variations and haplotypes in CaPV that linear reference alignments often miss.
• Comparative Population Structure: Provided the first detailed comparison of GTPV and SPPV population structures using graph-based metrics, revealing distinct evolutionary trajectories.
• ITR Plasticity Characterization: Mapped extensive structural rearrangements, truncations, and extended Open Reading Frames (ORFs) specifically at the Inverted Terminal Repeats (ITRs), linking them to host adaptation and attenuation.
• Standardized Nomenclature: Established a consistent gene mapping framework across GTPV, SPPV, and LSDV, facilitating the identification of pseudogenized or truncated genes in GTPV/SPPV relative to LSDV.

4. Key Results

A. Divergent Population Structures

• GTPV (Closed Pangenome): Exhibited a deep, stable population structure with three distinct clades (2.1, 2.2, 2.3).
  • Pangenome Status: Closed ($\alpha \approx 1.76$), indicating that additional sampling in Eurasia is unlikely to yield significant new mutations.
  • Genetics: Limited evidence of recent gene flow or admixture; clades are genetically stable and deeply divergent.
• SPPV (Open Pangenome): Showed a more recent, mosaic population structure with three clades (3.1, 3.2, 3.3) but weaker separation.
  • Pangenome Status: Open ($\alpha \approx 0.69$), suggesting that further sampling (especially from Africa) will reveal novel variation.
  • Genetics: Consistent with an ancestral bottleneck followed by recent population expansion.

B. Structural Variation at ITRs

• Haplotype Diversity: Structural variation was concentrated at the 5' and 3' ITRs.
  • GTPV: Clades 2.1/2.2 shared a 533 bp haplotype at the 5' ITR (LSDV002 region), while Clade 2.3 had a distinct 487 bp haplotype. Similar divergence was observed at the 3' ITR.
  • SPPV: Clade 3.1 (vaccine-related) possessed truncated, non-functional LSDV003 and LSDV154 genes. Clades 3.2/3.3 contained longer, functional isoforms.
• Extended ORFs: Several lineages showed extended putative ORFs spanning adjacent terminal genes, indicating recurrent structural plasticity. For example, SPPV Clades 3.2/3.3 had longer LSDV003 and LSDV154 isoforms with 124 additional amino acids.

C. Selection and Conservation

• Core vs. Terminal: The core genome (ORFs 24–123) was highly conserved, while the terminal regions showed high diversity.
• Selection Signals:
  • GTPV: Showed stronger purifying selection (lower $P_N/P_S$ ratio) and lower standard deviation in DoS values, consistent with older, stable lineages.
  • SPPV: Higher $P_N/P_S$ ratio and greater DoS variance, suggesting ongoing adaptive evolution.
  • Specific Loci: Genes LSDV004, LSDV023, LSDV068, and LSDV129 showed signals of adaptive evolution (DoS > 0.29).
  • Conserved Genes: 18 SPPV genes had zero or one SNP, suggesting strong functional constraints, whereas only one GTPV gene was similarly conserved.

D. Representative PVGs

• Constructed representative graphs for both species using genomes from major clades.
  • GTPV PVG: 152,917 bp, 6,865 nodes.
  • SPPV PVG: 151,016 bp, 1,611 nodes.
• These graphs serve as improved references for short-read mapping and variant detection.

5. Significance

• Evolutionary Insight: The study clarifies that GTPV evolution is driven by deep, ancient divergence, while SPPV evolution is characterized by recent expansion and structural plasticity.
• Host Specificity: The structural variations at the ITRs (involving genes like LSDV003, LSDV154, and LSDV026) are identified as key drivers of host range and virulence, potentially explaining why GTPV can infect sheep while SPPV is more restricted.
• Vaccine Surveillance: The identification of unique ITR mutations in vaccine strains (e.g., Clade 3.1 SPPV) highlights these regions as hotspots for recombination and attenuation, necessitating genomic surveillance of vaccine-derived strains.
• Methodological Advancement: This work validates PVGs as a superior tool for analyzing large DNA viruses, offering a framework to resolve complex structural variations that traditional linear reference methods cannot capture. It provides a critical resource for improving diagnostic accuracy and understanding the molecular basis of poxvirus host tropism.