JMIR Publications

ABSTRACT

Background:

Avian influenza virus reassortment is a major driver of viral evolution and the emergence of strains with altered pathogenicity, transmissibility, and zoonotic potential. Environmental surveillance has expanded as a population-level approach for influenza monitoring, yet inferring reassortment signals from environmental, pooled, and mixed samples remains challenging.

Objective:

This scoping review aimed to examine the computational methodologies used to infer and interpret influenza reassortment signals in environmental, pooled, or mixed samples.

Methods:

This scoping review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines. PubMed and Scopus were searched for studies applying computational methods to infer influenza reassortment in environmental, pooled, or mixed samples, with additional studies identified through citation tracking. Included studies were characterized by sample type, sequencing approach, and computational methodology used for reassortment inference.

Results:

Of 432 records screened, 48 studies were identified for this review. Most studies focused on low-complexity environmental samples, particularly fecal and surface swabs collected from live poultry markets and wild bird environments, whereas highly heterogeneous sample types such as municipal wastewater were mostly absent. All included studies relied on segment-wise phylogenetic incongruence as the primary basis for reassortment inference, with 56.3% supplementing tree comparisons using sequence similarity or genetic distance metrics. No studies applied formal statistical phylogenetic incongruence tests or dedicated reassortment-specific computational tools. Most workflows simplified sample complexity through virus isolation before sequencing, limiting their applicability to mixed viral populations.

Conclusions:

Current computational approaches for environmental influenza reassortment inference remain largely adapted from isolate-based analyses and are not well suited for complex mixed-population samples. Future surveillance efforts would benefit from computational methods capable of resolving mixed viral lineages, together with standardized reporting practices and sequencing workflows, to improve early detection of emerging reassortment signals.