Abstract
Abstract
In late 2025, influenza A(H3N2) Clade K viruses increased in frequency in global surveillance datasets. To characterize the molecular pattern associated with this emergence, we conducted genome-wide analysis of viruses collected during December 2025–January 2026 using geographically stratified sampling and a database-wide historical comparison with A(H3N2) sequences from 2019–2024. Predominant amino-acid substitutions were identified across multiple protein-coding open reading frames (ORF), including polymerase, surface, matrix, nucleoprotein, and nonstructural proteins. Historical database-wide analysis showed marked heterogeneity in the frequencies of these amino-acid states during 2019–2024, ranging from absent or rare to already highly prevalent. To examine their joint occurrence, polymorphic amino-acid states across the analyzed ORF were combined into genome-wide composite amino-acid signatures. Among 1,255 viruses with complete eight-segment sequence data, 168 distinct signatures were identified. A single signature predominated, occurring in 744 (59.3%) viruses, whereas the second most frequent signature occurred in 139 (11.1%); the remaining signatures were substantially less frequent. These findings demonstrate that the global emergence of influenza A(H3N2) Clade K was characterized by a predominant genome-wide amino-acid signature assembled from amino-acid states with markedly different historical frequencies. Genome-wide analysis of composite molecular signatures provides an additional framework for characterizing the genetic structure of rapidly expanding influenza virus populations.