Phylogenetic Analysis Traces Andes Virus Outbreak to Mid-20th Century Rodent Lineage

A genomic analysis of Andes virus (ANDV) sequences from the current MV Hondius-associated outbreak indicates the outbreak lineage likely originated from viral diversity circulating in long-tailed pygmy rice rats in Chile.

Molecular clock estimates place the closest sampled common ancestor of the outbreak lineage between the mid-1960s and late 1990s, depending on the genome segment analyzed. No evidence of elevated evolutionary rates or genetic reassortment was found in the outbreak lineage.

Methods

Researchers downloaded sequences from the European Nucleotide Archive (ENA), Pathoplexus, and GenBank, including eight sequences from the current outbreak.

Sequence alignments were performed using MAFFT v7.5. Molecular clock analyses were conducted in BEAST 2 v2.6.3 using HKY+I substitution models with uniform priors on clock rates. Ancestral sequence reconstruction was performed in IQ-TREE v2.3.5 using GTR+F+G4 models. Several sites were masked due to alignment artefacts.

Exclusion criteria for sequences included poor alignment, distant phylogenetic relatedness, and samples derived from cell-culture passage or experimental infection.

Results

Phylogenetic trees for the small (S), medium (M), and large (L) genome segments showed congruent topologies, supporting a rodent-reservoir origin for the outbreak. No evidence of reassortment among the sampled lineages was detected.

The sampled outbreak diversity is consistent with sequence WDKH.1 (collected 26 April 2026) as a potential root. An alternative rooting at position M:3500 in sequence W6RC.2 requires raw-read validation.

• M segment: The closest sampled common ancestor of the outbreak lineage was estimated to May 1999 (95% highest posterior density [HPD]: February 1985 – July 2007) in one analysis, and April 1998 (95% HPD: September 1982 – May 2007) in another.
• S segment (rodent-inclusive): Estimated to July 1964 (95% HPD: December 1879 – February 1995) in one analysis, and March 1959 (95% HPD: December 1847 – September 1995) in another.
• L segment: No rodent-inclusive estimate was possible; a human-only analysis yielded a more recent date.

Researchers noted that date estimates reflect the closest sampled common ancestor of the outbreak lineage, not the precise timing of zoonotic spillover.

Substitution rate estimates were timescale-dependent: the inclusion of older sequences in the analysis lowered rate estimates and pushed back common ancestor dates.

Most coding-region mutations on the stem branch (the lineage leading to the outbreak) were nonsynonymous:

• S segment: 11 of 13 mutations
• M segment: 16 of 18 mutations
• L segment: 45 of 49 mutations

A one-sided binomial test indicated a statistically significant enrichment of nonsynonymous mutations in the L segment (p = 0.0096; Bonferroni-corrected p = 0.0288). No elevated substitution rate was detected on the stem branch.

Interpretation

The analysis supports a rodent reservoir origin for the outbreak and places the closest sampled common ancestor to the outbreak lineage in the late 1990s based on M segment estimates. No elevated molecular clock rate was detected, but the excess of nonsynonymous mutations in the L segment warrants further investigation, according to the researchers.

Contributors

Charu Sharma, Sanni Översti, Katrina Lythgoe, Aris Katzourakis (University of Oxford); Spyros Lytras (Institut Pasteur); Mahan Ghafari (University of Oxford).