The study, led by Adrian S. Arcillo and Maria Llaine J. Callanta of the University of the Philippines Manila together with collaborators from Philippine government research institutions, assembled every publicly available Reston virus whole-genome sequence and subjected it to a battery of phylogenetic and population genetic analyses. After rigorous filtering of the NCBI Virus database, which initially yielded 66 nucleotide sequences, the team retained 20 complete genomes meeting strict quality criteria: each had to be at least 18.8 kilobase pairs long and contain all seven canonical filovirus coding regions, from the nucleoprotein gene through to the large RNA-dependent RNA polymerase. Unverified, patent-derived, synthetic, partial and experimentally reinfection-derived sequences were excluded to prevent artifacts from contaminating downstream evolutionary inference.

Reston virus, formally Orthoebolavirus restonense, belongs to the family Filoviridae, the same family as Ebola virus and Marburg virus. It is a non-segmented, negative-sense RNA virus whose genetic variation arises primarily through errors during RNA replication, driven by an error-prone polymerase and limited proofreading capacity. Because recombination can bias phylogenetic reconstruction, the researchers first screened the curated alignment using seven independent detection methods implemented in RDP5. No recombination event met the significance threshold, a finding consistent with the generally conserved genome organization of filoviruses and reassuring for the validity of the subsequent analyses.

Pairwise comparison of the 20 genomes revealed sequence identities ranging from 95.22 to 100 percent, indicating substantial genomic conservation within the species while still preserving enough nucleotide differences to resolve outbreak-associated lineages. Maximum-likelihood phylogenetic reconstruction, performed under the best-fit substitution model TVM+F+G4 with 1,000 bootstrap replicates, supported five multi-sequence clusters and two phylogenetically distinct singleton lineages, all with bootstrap support exceeding 90 percent. The lineages mapped cleanly onto epidemiological history: primate-derived isolates from the 1989–1990 outbreaks in Virginia and Pennsylvania formed one lineage, monkey-derived isolates from Siena in 1992 and Texas in 1996 formed two more, swine-derived isolates from the 2008–2009 Philippine outbreaks formed additional lineages, and monkey-derived isolates from the 2015 Philippine re-emergence formed the final cluster.

The most consequential result emerged from the Bayesian molecular-clock analysis. Before running the clock, the team verified temporal signal using TempEst, obtaining a correlation coefficient of 0.992 and a coefficient of determination of 0.984, meaning that 98.4 percent of root-to-tip genetic divergence was explained by sampling date. Under a relaxed lognormal clock and a coalescent constant-population model in BEAST2, the time to the most recent common ancestor of the sampled genomes was estimated at approximately 1967, with a broad 95 percent highest posterior density interval spanning 1930 to 1988. The lineage associated with the first recorded outbreak in Reston, Virginia coalesced around 1987, two years before its identification. Because the monkeys involved in the early outbreaks were traced to breeding and export facilities in the Philippines, the pre-1989 ancestry, combined with the divergent placement of certain swine isolates, raises the possibility of preexisting Reston virus circulation in the country, though the responsible hosts and continuity of transmission remain unresolved.

Beyond phylogeny, the study quantified genetic diversity and population structure. The pooled dataset showed high haplotype diversity, with 17 unique haplotypes among just 20 genomes. Stratifying by host revealed that swine-derived genomes exhibited greater nucleotide diversity than primate-derived genomes, a pattern the authors interpret as indicative of a prolonged evolutionary history, an admixture of historically differentiated lineages, or both. Primate-derived sequences, by contrast, showed high haplotype diversity but low nucleotide diversity, a signature of rapid population growth following a bottleneck or founder effect, comparable to patterns observed in Ebola virus during the 2014–2016 West African epidemic.

Population structure analyses told a more nuanced story. Both host-based and temporal stratifications produced high FST values exceeding 0.25, consistent with marked genetic differentiation. However, analysis of molecular variance attributed 52.4 percent of genetic variation to differences among outbreak-period groups versus 47.6 percent within periods, while host-based stratification attributed only 38.4 percent to differences between primate and swine populations. The nearest-neighbor statistic Snn yielded high values of 0.900 under both schemes, confirming non-random sequence clustering. Taken together, these results indicate that outbreak period exerted a stronger influence on the observed genetic structure than host association, although the authors caution that the relative contributions of host and time cannot be fully disentangled because each outbreak period was predominantly associated with a particular host.

Neutrality tests based on Tajima’s D and Fu’s Fs provided no evidence of recent rapid population expansion. All Tajima’s D values were non-significant, while Fu’s Fs yielded consistently high positive values across all analyses, a pattern the authors suggest may reflect a decrease of haplotypes relative to neutral expectations, potentially arising from previous population bottlenecks, founder events, or genetic structuring associated with independent viral introductions into monkeys and swine. The combination of high haplotype and nucleotide diversity alongside no detectable demographic expansion supports the idea of an established Reston virus population rather than a recently emerged one.

The authors are careful to acknowledge the limitations imposed by their dataset. Twenty genomes, however rigorously curated, constrain the statistical power of diversity, neutrality and population structure analyses, particularly after stratification by host and outbreak period. Spatial stratification was not attempted because geographic metadata from GenBank records and associated literature were incomplete and inconsistently resolved. The 2015 outbreak group, represented by only two sequences, could not support meaningful neutrality estimates. These constraints mean the results must be interpreted as a baseline framework rather than a definitive account of Reston virus evolutionary history.

Nevertheless, the study’s implications for surveillance are clear. Existing Reston virus genomic data are largely outbreak-driven, with several detections made incidentally during diagnostic investigations of other infections, such as the 2008–2009 swine outbreak discovered through co-infection with porcine reproductive and respiratory syndrome virus. Molecular and serological evidence from Philippine bats, including PCR-positive but antibody-negative individuals, suggests active infection prior to antibody development, while seropositive but PCR-negative animals indicate prior exposure. Experimental studies in Marburg virus demonstrate that viral RNA shedding is brief whereas antibody responses endure, a pattern mirrored in Reston virus-infected swine and non-human primates. Together, these findings suggest that Reston virus circulation may be overlooked when surveillance relies on a single diagnostic approach or is limited to recognized outbreaks. The authors argue that a One Health surveillance framework linking animal, human and environmental monitoring, complemented by active molecular detection and serological monitoring, is essential to capture the full extent of Reston virus diversity, elucidate its zoonotic potential, and strengthen preparedness for future changes in its epidemiology.

Subject of Research: Whole-genome phylogenetic and population genetic analysis of Orthoebolavirus restonense

Article Title: Whole-genome phylogenetic and population genetic analysis of Orthoebolavirus restonense

Article References: Arcillo, A. S., Collado, A. R. G., Malabad, J. C. M., Miranda, N. L. J., Villavicencio, A. G. M., Ong, R. A. K., Guevarra, R. F. I., & Callanta, M. L. J. (2026). Whole-genome phylogenetic and population genetic analysis of Orthoebolavirus restonense. Frontiers in Tropical Diseases, 7, Article 1955687. https://doi.org/10.3389/fitd.2026.1955687

Image Credits: AI Generated

DOI: 10.3389/fitd.2026.1955687

Keywords: Whole-genome, phylogenetic, population, genetic, analysis, Orthoebolavirus, restonense, scientific research