Metadata completeness and phylogenetic structure of publicly available Streptococcus anginosus 16 S rRNA gene sequences

Streptococcus anginosus has recently emerged as a candidate gastric pathogen implicated in inflammation and tumorigenesis, prompting evaluation of whether public 16 S rRNA gene archives support phylogeographic or epidemiological analyses. We analyzed 350 S. anginosus 16S rRNA gene sequences (1,200–1,700 bp) that were deposited in the NCBI database and manually curated host, geographic, and sample-source metadata. Sequences were screened for taxonomic and compositional outliers using SILVA-based identity and composition-homogeneity tests, yielding a curated set of 346 sequences. Both datasets were aligned with MAFFT and used for maximum-likelihood phylogenetic inference in IQ-TREE, with ModelFinder-selected models and 1,000 ultrafast bootstrap replicates. Geographic clustering was assessed using a topology-based Fitch parsimony permutation test. All 350 records were reported as S. anginosus. Four records failed the composition-homogeneity test; two also had top SILVA hits outside Streptococcus, and all four were excluded from the curated dataset. Metadata were frequently incomplete, with missing sample source (30.9%), geographic origin (21.7%), and host (16.6%). The parsimony permutation test showed fewer country-of-origin state changes than expected by chance in both the curated 346-sequence tree (62.0% of null mean; p < 0.0001) and full 350-sequence tree (63.4%; p < 0.0001). This pattern was robust to the exclusion of the largest study and across 1,000 bootstrap trees but likely reflects study-specific sampling rather than confirmed phylogeographic structure. Public S. anginosus 16S rRNA datasets are abundant but poorly annotated. Improved metadata standards and multilocus or whole-genome sequencing are needed for robust epidemiological and strain-level investigations.