Under Construction: Classification of Hepatitis B Virus Genotypes and Subgenotypes

(2) Sequencing of only parts of the HBV genomes led to the tentative suggestion of subgenotypes A4 and A5 [20] . In an article in Emerging Infectious Diseases sequencing whole genomes of tentative A4 isolates, Andernach et al. [21] now group this subgenotype into the clade of A5. (3) Based on whole genome sequencing two new subgenotypes have been called C6 in two contemporaneous publications. C6 has been identified in the Philippines [22] and Papua, Indonesia [23] . Isolates from these two publications map into two clades with a divergence of 5.1% and thus make up a new subgenotype C7 from the Philippines [Schaefer S, unpublished]. (4) A similar problem seems to arise with subgenotype D6. There are two concurrent publications that describe a new subgenotype D6 in Tunisia [24] and Papua, Indonesia [23] . (5) Regarding the delimitation of subgenotypes, these have been described as the major divisions within the phylogenetic trees in genotypes by Norder et al. [4] . Later on Kramvis et al. [14] defined subgenotypes as differing by at least 4% and supported this by robust bootstrapping data, although yielding largely overlapping results with the definition by Norder et al. [4] . This competing definition is partly contradictory to the original paper [4] , where 4 major clades of genotype D, designated D1–D4, were described with an overall divergence for some of these clearly below 4% [25] . However, these subgenotypes Hepatitis B virus (HBV) genotypes [1–6] and subgenotypes [7–12] have an impact on the course of disease and treatment decisions. In Germany, HBV genotyping is recommended before starting interferon therapy treatment of HBV infection [13] . Thus, an accurate and universal system of classification is of prime importance. Several largely overlapping definitions for HBV genotypes have been presented by leading experts in this field in the past [4, 14–17] . However, due to the lack of universally accepted rules, irregularities have accumulated within the last 20 years of HBV genotype research. Some of these are found below. (1) In a letter to the editor in this issue of Intervirology Ahn et al. (pp. 321–322) raise an important point. The assignation of subgenotypes C1 and C2 to clades is contradictory in the literature. As outlined by Ahn et al. there are two influential papers in which C1 following the nomenclature of Norder et al. [4] is C2 according to the classification of Huy et al. [18] and vice versa. However, regarding the figure in the letter, an uninitiated reader might get the impression that Huy et al. [18] also defined subgenotypes C3–C5. Also, only C1–C4 were described by Norder et al. [4] . C5 was described by Sakamoto et al. [19] in 2006. The reason for the discrepancies in nomenclature would have been more obvious if the references in the figure had been Huy et al. [18] and Norder et al. [4] , respectively. Published online: September 29, 2009