2.7. HBV genetic compartmentalization tests

Compartmentalization analysis of serum and CSF samples from each patient was performed using HyPhy software (V2.5).¹⁵ As different methods to define genetic compartmentalization (such as genetic heterogeneity and distinct lineages between two viral subpopulations) might occasionally yield discordant results in compartmentalization tests,¹⁶ we applied genetic distance-based tests and tree-based tests for the genetic compartmentalization analysis. The distance-based tests included Fst and Snn (nearest-neighbor statistic), and Fst has several different types: 1) Hudson, Slatkin and Madison (HSM), 2) Slatkin (S), and 3) Hudson, Boos and Kaplan (HBK). We employed the HBK nonparametric test for the distance-based test and computed the K_ST statistic, whose values range from 0 (denoting no compartmentalization) to 1 (denoting complete compartmentalization). In addition, the Slatkin-Maddison (SM) method was used for the tree-based test. A p lower than 0.05 indicates that the elements can be judged to be compartmentalized. Compartmentalization was finally confirmed by two algorithms (K_ST and SM tests) when both p values were less than 0.05.