# Constraint of Base Pairing on HDV Genome Evolution

^{*}

## Abstract

**:**

## 1. Introduction

^{−3}per nucleotide site per year. Chao, Tang, and Hsu (1994) found that the evolutionary rates of different parts of the HDV genome vary [26]. Krushkal and Li (1995) estimated the evolutionary rates of the coding and non-coding region of HDV [27]. In their study, the synonymous and non-synonymous evolutionary rates were estimated for the coding region. The orders of the estimated evolutionary rates are the same as those of previous studies. The synonymous evolutionary rates are smaller than the non-synonymous evolutionary rates and the evolutionary rates of the non-coding region. They considered that the strong GC bias at the third codon position is the cause of the low rates of the synonymous substitutions. Wu et al. (1999) identified nine segments of the HDV genome with high levels of recombination [28]. They suggested that the recombination can contribute to the genetic diversity of HDV. Anisimova and Yang (2004) examined 33 sequences of the HDAg gene from Genotypes I, II, and III to study the diversifying selection on the HDAg gene [29]. Then, it was found that about 11% codon sites of the HDAg gene are under positive selection. No significant evidence for recombination was obtained from their study. Bishal, Mukherjee, and Chakraborty (2013) investigated the synonymous codon usage pattern of the HDAg gene to find that the most important determinant of the codon usage is the mutation pressure [30]. Delfino et al. (2018) reported comprehensive computational analysis of HDV full-length genomes, in which phylogenetic relationships and various characteristics of the amino acid sequences are described [8].

## 2. Materials and Methods

#### 2.1. Alignment of HDV Genomes and L-HDAg Genes

#### 2.2. Threshold for Base-Pairing Probability

#### 2.3. Calculation of Evolutionary Distance

_{bp}(x, y) and ed

_{nbp}(x, y).

#### 2.4. Analysis of Distance Ratio

_{a}and v

_{b}, and that a pair of aligned sequences, x and y, diverged from the common ancestral sequence at time t

_{ab}. For simplicity, let us assume that the evolutionary rates did not change during the time after the divergence from the ancestral sequence. Then, the evolutionary distances of regions a and b between x and y, d

_{a}(x, y) and d

_{b}(x, y), are expressed as follows (see Figure 2a):

_{bp}(x, y)/ed

_{nbp}(x, y), calculated between every possible pair of aligned sequences except for the pairs with the distance ed

_{nbp}(x, y) = 0.0, is expected to take a value close to one. The one-sample Wilcoxon test can be applied to the data of the distance ratios, to examine whether the base-pairing functions as the constraint on the base-pairing sites (see Figure 2c). The null hypothesis of the test is that the median of the data = 1.0. If the base-pairing function is the constraint, the evolutionary rate of the base-pairing sites would be smaller than that of the non-base-pairing sites. So, the alternative hypothesis is set that the median of the data < 1.0.

_{bp}(x, y)/ed

_{nbp}(x, y), nine other distance ratios, ed

_{bp}(x, y)/ed(x, y), ed

_{nbp}(x, y)/ed(x, y), ed

_{bp}(x, y)/${K}_{s}^{c}\left(x,y\right)$, ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right),$ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right),$ ed

_{bp}(x, y)/${K}_{a}^{c}\left(x,y\right)$, ed

_{nbp}(x, y)/${K}_{a}^{c}\left(x,y\right),$ed

_{nbp}(x, y)/${K}_{a}^{c}\left(x,y\right),$and ${K}_{a}^{c}\left(x,y\right)$/${K}_{s}^{c}\left(x,y\right)$, were examined by the one-sample Wilcoxon test. The setting of the alternative hypothesis for each distance ratio is described in the legends for Table 3 and Table 4. The computational language for statistical analysis, R [44], was used for the one-sample Wilcoxon test. The effect of the multiple comparison on the p-values was corrected by the Bonferroni method. That is, the corrected p-values were obtained by multiplying the p-values by ten. The corrected p-values were examined with the significance level = 0.01. The distances within each of three Genotypes and those between every possible pair of the Genotypes were calculated, and the ten distance ratios were analyzed by the procedure described above.

## 3. Results

#### 3.1. Comparison within Each Genotype

_{bp}(x, y), ed

_{nbp}(x, y), ed(x, y), ${K}_{s}^{c}\left(x,y\right),$ and ${K}_{a}^{c}\left(x,y\right),$between every possible pair of aligned sequences, except for the pair which has ed

_{nbp}(x, y) = 0.0, ed(x, y) = 0.0, ${K}_{s}^{c}\left(x,y\right)$ = 0.0, or ${K}_{s}^{c}\left(x,y\right)$ = 0.0, are shown in Supplementary Data S3, and the summary of the distances is shown in Supplementary Data S4. The pairs with ed

_{nbp}(x, y), ed(x, y), ${K}_{s}^{c}\left(x,y\right)$, or ${K}_{a}^{c}\left(x,y\right)$ = 0.0 were excluded from the analysis, because the four distances were used as the denominators of the distance ratios. The numbers of base-pairing sites, non-base-pairing sites, the entire non-coding region, synonymous sites, and non-synonymous sites are also shown in Supplementary Data S3. The box plots of the ten distance ratios calculated by the sequence comparison within each genotype are shown in Figure 3. As shown in the figure, the distance ratios of either genotype did not seem to follow the normal distribution. So, the distance ratios were examined by the one-sample Wilcoxon test (see Table 3).

_{bp}(x, y)/ed

_{nbp}(x, y), ed

_{bp}(x, y)/ed(x, y), ed

_{nbp}(x, y)/ed(x, y), ed

_{bp}(x, y)/${K}_{s}^{c}\left(x,y\right)$, ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right),$ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right),$ ed

_{bp}(x,y)/${K}_{a}^{c}\left(x,y\right)$, ed

_{nbp}(x,y)/${K}_{a}^{c}\left(x,y\right),$ed

_{nbp}(x, y)/${K}_{a}^{c}\left(x,y\right),$ and ${K}_{a}^{c}\left(x,y\right)$/${K}_{s}^{c}\left(x,y\right),$calculated by the sequence comparison within each genotype are shown in Table 3. All the corrected p-values for the ten distance ratios were less than 0.01 for either Genotype I or II. In contrast, all the corrected p-values of Genotype III were greater than 0.01, which may be due to the small sample size (see Table 3). Therefore, only the distance ratios within Genotype I and those within Genotype II were examined. Out of the ten distance ratios, we focused on the six ratios, ed

_{bp}(x, y)/ed

_{nbp}(x, y), ed(x, y)/${K}_{s}^{c}\left(x,y\right),$ ed

_{bp}(x, y)/${K}_{s}^{c}\left(x,y\right)$, ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right)$, ed

_{bp}(x, y)/${K}_{a}^{c}\left(x,y\right)$, and ed

_{nbp}(x, y)/${K}_{a}^{c}\left(x,y\right)$, to examine the effect of the base pairing.

_{bp}(x, y)/ed

_{nbp}(x, y) is equal to 1.0 was rejected for either Genotype I or II. The alternative hypothesis was that the median of ed

_{bp}(x, y)/ed

_{nbp}(x, y) is less than 1.0. The results suggest that the constraint on the base-pairing sites is stronger than that on non-base-pairing sites. That is, base-pairing functions as the constraint on the non-coding region.

_{bp}(x, y)/${K}_{s}^{c}\left(x,y\right)$ and ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right)$, were examined by the one-sample Wilcoxon test. The null hypothesis that the median of ed

_{bp}(x, y)/${K}_{s}^{c}\left(x,y\right)$ is equal to 1.0 was rejected for either Genotype I or II. The alternative hypothesis was that ed

_{bp}(x, y)/${K}_{s}^{c}\left(x,y\right)$ is less than 1.0. That is, the constraint on the base-pairing sites is stronger than that on the synonymous sites. The null hypothesis that the median of ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right)$ is equal to 1.0 was also rejected. The alternative hypothesis was that the median of ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right)$ is less than 1.0. That is, the constraint on the non-base-pairing sites is stronger than that on the synonymous sites.

_{bp}(x, y)/${K}_{a}^{c}\left(x,y\right)$ and ed

_{nbp}(x, y)/${K}_{a}^{c}\left(x,y\right)$. The null hypothesis that the median of ed

_{bp}(x, y)/${K}_{a}^{c}\left(x,y\right)$ is equal to 1.0 was rejected for either Genotype I or II. The alternative hypothesis was that ed

_{bp}(x, y)/${K}_{a}^{c}\left(x,y\right)$ is less than 1.0. This means that the constraint on the non-coding region is stronger than that on the non-synonymous sites. The null hypothesis that the median of ed

_{nbp}(x, y)/${K}_{a}^{c}\left(x,y\right)$ is equal to 1.0 was also rejected. The alternative hypothesis was that ed

_{nbp}(x, y)/${K}_{a}^{c}\left(x,y\right)$ is greater than 1.0. The results suggest that the constraint on the non-base-pairing sites is weaker than that on the non-synonymous sites.

#### 3.2. Comparison between Genotypes

_{nbp}(x, y)/ed(x, y), between Genotypes I and II were less than $2.2\times {10}^{-15}$. The corrected p-value corresponding to ed

_{nbp}(x, y)/ed(x, y) between Genotypes I and II was 1.8$\times {10}^{-7}$. So, all the corrected p-values were less than 0.01. The results of the statistical analyses for the distance ratios between Genotypes were basically consistent with those within Genotype I or II. Out of the ten distance ratios, we here describe the results of the statistical analyses, focusing on the six ratios, ed

_{bp}(x, y)/ed

_{nbp}(x, y), ed(x, y)/${K}_{s}^{c}\left(x,y\right),$ ed

_{bp}(x, y)/${K}_{s}^{c}\left(x,y\right)$, ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right)$, ed

_{bp}(x, y)/${K}_{a}^{c}\left(x,y\right)$, and ed

_{nbp}(x, y)/${K}_{a}^{c}\left(x,y\right)$, to examine the effect of the base pairing.

_{bp}(x, y)/ed

_{nbp}(x, y) = 1.0 was rejected. The alternative hypothesis was that the median of ed

_{bp}(x, y)/ed

_{nbp}(x, y) < 1.0. The results suggest that the constraint on the base-pairing sites is stronger than that on non-base-pairing sites.

_{bp}(x, y)/${K}_{s}^{c}\left(x,y\right)$ and ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right)$, were examined. The corrected p-values for the null hypothesis that the median of ed

_{bp}(x, y)/${K}_{s}^{c}\left(x,y\right)=1.0$ was less than 0.01 for either pair of the genotypes. The alternative hypothesis was that the median of the distance ratios is less than 1.0. This means that the constraint on the base-pairing sites is greater than that on the synonymous sites. The corrected p-values for the null hypothesis that the median of ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right)$ is equal to 1.0 were less than 0.01, for either pair of the genotypes. The alternative hypothesis was that the median of ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right)$ is less than 1.0. So, not only the constraint on base-pairing sites, but also that on the non-base-pairing sites, is stronger than that on the synonymous sites.

_{bp}(x, y)/${K}_{a}^{c}\left(x,y\right)$ is equal to 1.0 were less than 0.01 for either pair of the genotypes. The alternative hypothesis was that the median is less than 1.0. The result suggests that the constraint on the base-pairing sites is stronger than that on the non-synonymous sites. The null hypothesis that the median of ed

_{nbp}(x, y)/${K}_{a}^{c}\left(x,y\right)$ is equal to 1.0 was also examined. The corrected p-value was less than 0.01 for either pair of the Genotypes. The alternative hypothesis for the ratios between Genotypes I and II was that the median of the ratios is less than 1.0. That is, the constraint on the non-base-pairing sites is stronger than that on the non-synonymous sites. In contrast, the alternative hypothesis for ratios between other pairs of the Genotypes was that the median is greater than 1.0. The setting of the latter alternative hypothesis was the same as that for the ratios within Genotype I or II. The results suggest that the constraint on the non-base-pairing sites is weaker than that on the non-synonymous sites.

## 4. Discussion

_{bp}(x, y)/ed

_{nbp}(x, y), obtained from the sequence comparison within a genotype and those between a pair of genotypes were examined by the one-sample Wilcoxon test. The results suggest that the constraint on the base-pairing sites is stronger than that on the non-base-pairing sites in the non-coding region. That is, the base pairing is considered to function as a constraint on the HDV genome.

_{bp}(x, y)/${K}_{s}^{c}\left(x,y\right)$ and ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right)$, were examined. The results suggest that not only the constraint on the base-pairing sites but also that on the non-base-pairing sites is stronger than that on the synonymous sites. The cause of constraint on the non-base-pairing sites in the non-coding region is unknown. It is suggested that the disrupted region of secondary structure, such as mismatches, bulges, and loops, may be involved in the arrangement of adenosine deaminase, the RNA editing enzyme, toward the editing site [18]. Therefore, the non-base-pairing sites may have some functional roles, which would cause the constraint on the nucleotide substitution in the non-base-pairing set. The medians of three distance ratios, ed

_{bp}(x, y)/${K}_{s}^{c}\left(x,y\right)$, ed

_{nbp}(x, y)/${K}_{s}^{c}\left(x,y\right)$, and ${K}_{a}^{c}\left(x,y\right)$/${K}_{s}^{c}\left(x,y\right),$ were compared. The order relation of the medians of the three distance ratios was identical among the comparisons within Genotype I, within Genotype II, between Genotypes I and III, and between Genotypes II and III.

_{bp}(x, y)/${K}_{a}^{c}\left(x,y\right)$ and ed

_{nbp}(x, y)/${K}_{a}^{c}\left(x,y\right)$ and the medians were consistent with the relation. In contrast, the order relation of distance ratios obtained from the comparison between Genotypes I and II was different from those described above.

_{bp}(x, y)/${K}_{a}^{c}\left(x,y\right)$ ed

_{nbp}(x, y)/${K}_{a}^{c}\left(x,y\right)$ and the medians were consistent with the order relation.

## Supplementary Materials

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Husa, P.; Linhartová, A.; Nemecek, V.; Husová, L. Hepatitis D. Acta Virol.
**2005**, 49, 219–225. [Google Scholar] - Sureau, C.; Guerra, B.; Lanford, R.E. Role of the large hepatitis B virus envelope protein. J. Virol.
**1993**, 67, 366–372. [Google Scholar] [CrossRef][Green Version] - Sureau, C. The role of the HBV envelope proteins in the HDV replication cycle. Curr. Top. Microbiol. Immunol.
**2006**, 307, 113–131. [Google Scholar] [CrossRef] - Niro, G.A.; Smedile, A.; Andriulli, A.; Rizzetto, M.; Gerin, J.L.; Casey, J.L. The predominance of hepatitis delta virus genotype I among chronically infected Italian patients. Hepatology
**1997**, 25, 728–734. [Google Scholar] [CrossRef] - Shakil, A.O.; Hadziyannis, S.; Hoofnagle, J.H.; Di Bisceglie, A.M.; Gerin, J.L.; Casey, J.L. Geographic distribution and genetic variability of hepatitis delta virus genotype I. Virology
**1997**, 234, 160–167. [Google Scholar] [CrossRef] [PubMed][Green Version] - Dény, P. Hepatitis delta virus genetic variability: From genotypes I, II, III to eight major clades? Curr. Top. Microbiol. Immunol.
**2006**, 307, 151–171. [Google Scholar] [CrossRef] - Le Gal, F.; Gault, E.; Ripault, M.-P.; Serpaggi, J.; Trinchet, J.-C.; Gordien, E.; Dény, P. Eighth major clade for hepatitis delta virus. Emerg. Infect. Dis.
**2006**, 12, 1447–1450. [Google Scholar] [CrossRef] [PubMed] - Delfino, C.M.; Cerrudo, C.S.; Biglione, M.; Oubiña, J.R.; Ghiringhelli, P.D.; Mathet, V.L. A comprehensive bioinformatics analysis of hepatitis D virus full-length genomes. J. Viral. Hepat.
**2018**, 25, 860–869. [Google Scholar] [CrossRef] [PubMed][Green Version] - Wille, M.; Netter, H.J.; Littlejohn, M.; Yuen, L.; Shi, M.; Eden, J.-S.; Klaassen, M.; Holmes, E.C.; Hurt, A.C. A divergent hepatitis D-like agent in birds. Viruses
**2018**, 19, 720. [Google Scholar] [CrossRef] [PubMed][Green Version] - Chang, W.-S.; Pattersson, J.H.; Le Lay, C.; Shi, M.; Lo, N.; Wille, M.; Eden, J.-S.; Holmes, E.C. Novel hepatitis D-like agents in vertebrates and invertebrates. Virus Evol.
**2019**, 5, vez021. [Google Scholar] [CrossRef] - Wang, K.S.; Choo, Q.L.; Weiner, A.J.; Ou, J.H.; Najarian, R.C.; Thayer, R.M.; Mullenbach, G.T.; Denniston, K.J.; Gerin, J.L.; Houghton, M. Structure, sequence, and expression of the hepatitis delta (δ) viral genome. Nature
**1986**, 323, 508–514. [Google Scholar] [CrossRef] - Taylor, J.M. The structure and replication of hepatitis delta virus. Annu. Rev. Microbiol.
**1992**, 46, 253–276. [Google Scholar] [CrossRef] - Elena, S.F.; Dopazo, J.; Flores, R.; Diener, T.O.; Moya, A. Phylogeny of viroids, viroidlike satellite RNAs, and viroidlike domain of hepatitis d virus RNA. Proc. Natl. Acad. Sci. USA
**1991**, 88, 5631–5634. [Google Scholar] [CrossRef][Green Version] - Jenkins, G.M.; Woelk, C.H.; Rambaut, A.; Holmes, E.C. Testing the extent of sequence similarity among viroids, satellite RNAs, and hepatitis delta virus. J. Mol. Evol.
**2000**, 50, 98–102. [Google Scholar] [CrossRef] [PubMed] - Sharmeen, L.; Kuo, M.Y.P.; Dinter-Gottlieb, G.; Taylor, J. Antigenomic RNA of human hepatitis delta virus can undergo self-cleavage. J. Virol.
**1988**, 62, 2674–2679. [Google Scholar] [CrossRef] [PubMed][Green Version] - Ferré-D’Amaré, A.R.; Zhou, K.; Doudna, J.A. Crystal structure of a hepatitis delta virus ribozyme. Narture
**1998**, 395, 567–574. [Google Scholar] [CrossRef] - Been, M.D. HDV ribozyme. Curr. Top. Microbiol. Immunol.
**2006**, 307, 67–89. [Google Scholar] [CrossRef] - Casey, J.L. RNA editing in hepatitis delta virus. Curr. Top. Microbiol. Immunol.
**2006**, 307, 67–89. [Google Scholar] [CrossRef] - Kuo, M.Y.P.; Chao, M.; Taylor, J. Initiation of replication of the human hepatitis delta virus genome from cloned DNA: Role of delta antigen. J. Virol.
**1989**, 63, 1945–1950. [Google Scholar] [CrossRef][Green Version] - Chao, M.; Hsieh, S.Y.; Taylor, J. Role of two forms of hepatitis delta virus antigen: Evidence for a mechanism of self-limiting genome replication. J. Virol.
**1990**, 64, 5066–5069. [Google Scholar] [CrossRef][Green Version] - Glenn, J.S.; White, J.M. Trans-dominant inhibition of human hepatitis delta virus genome replication. J. Virol.
**1991**, 65, 2357–2361. [Google Scholar] [CrossRef] [PubMed][Green Version] - Chang, F.L.; Chen, P.L.; Tu, S.J.; Wang, G.J.; Chen, D.S. The large form of hepatitis delta antigen is crucial for assembly of hepatitis delta virus. Proc. Natl. Acad. Sci. USA
**1991**, 88, 8490–8494. [Google Scholar] [CrossRef] [PubMed][Green Version] - Ryu, W.S.; Bayer, M.; Taylor, J. Assembly of hepatitis delta virus particles. J. Virol.
**1992**, 66, 2310–2315. [Google Scholar] [CrossRef] [PubMed][Green Version] - Imazeki, F.; Omata, M.; Ohto, M. Heterogeneity and evolution rates of delta virus RNA sequences. J. Virol.
**1990**, 64, 5594–5599. [Google Scholar] [CrossRef] [PubMed][Green Version] - Lee, C.-M.; Bih, F.-Y.; Chao, Y.-C.; Govindarajan, S.; Lai, M.M.C. Evolution of hepatitis delta virus RNA during chronic infection. Virology
**1992**, 188, 265–273. [Google Scholar] [CrossRef] - Chao, Y.-C.; Tang, H.-S.; Hsu, C.-T. Evolution rate of hepatitis delta virus RNA isolated in Taiwan. J. Med. Virol.
**1994**, 43, 397–403. [Google Scholar] [CrossRef] [PubMed] - Krushkal, J.; Li, W.-H. Substitution rates in hepatitis D virus. J. Mol. Evol.
**1995**, 41, 721–726. [Google Scholar] [CrossRef] - Wu, J.C.; Chiang, T.Y.; Shiue, W.K.; Wang, S.Y.; Sheen, I.J.; Huang, Y.H.; Syu, W.J. Recombination of hepatitis D virus RNA sequences and its implications. Mol. Biol. Evol.
**1999**, 16, 1622–1632. [Google Scholar] [CrossRef] - Anisimova, M.; Yang, Z. Molecular evolution of the hepatitis delta virus gene: Recombination or positive selection. J. Mol. Evol.
**2004**, 59, 815–826. [Google Scholar] [CrossRef][Green Version] - Bishal, A.K.; Mukherjee, R.; Chakraborty, C. Synonymous codon usage pattern analysis of hepatitis D virus. Virus Res.
**2013**, 173, 350–353. [Google Scholar] [CrossRef] [PubMed] - Katoh, K.; Standley, D.M. MAFFT multiple sequence alignment software version 7: Improvements in performance and usability. Mol. Biol. Evol.
**2013**, 30, 772–780. [Google Scholar] [CrossRef] [PubMed][Green Version] - Katoh, K.; Toh, H. Improved accuracy of multiple ncRNA alignment by incorporating structural information into a MAFFT-based framework. BMC Bioinform.
**2008**, 9, 212. [Google Scholar] [CrossRef][Green Version] - McCaskill, J.S. The equilibrium partition function and base pair binding probabilities for RNA secondary structure. Biopolymers
**1990**, 28, 1105–1119. [Google Scholar] [CrossRef] [PubMed] - Hofacker, I.L.; Fekete, M.; Stadler, P.F. Secondary structreu prediction for aligned RNA sequences. J. Mol.Biol.
**2002**, 319, 1059–1066. [Google Scholar] [CrossRef] - Rice, P.; Longden, I.; Bleasby, A. EMBOSS: The European Molecular Biology Open Software Suite. Trends Genet.
**2000**, 16, 276–277. [Google Scholar] [CrossRef] - Saitou, N.; Nei, M. The neighbor-joining method: A new method for reconstructing phylogenetic trees. Mol. Biol. Evol.
**1987**, 4, 406–425. [Google Scholar] [CrossRef] [PubMed] - Kumar, S.; Stecher, G.; Li, M.; Knyaz, C.; Tamura, K. MEGA X: Molecular evolutionary genetics analysis across computing platforms. Mol. Biol. Evol.
**2018**, 35, 1547–1549. [Google Scholar] [CrossRef] - Imazeki, F.; Omata, M.; Ohto, M. Complete nucleotide sequence of hepatitis delta virus RNA in Japan. Nucl. Acid. Res.
**1991**, 19, 5439–5440. [Google Scholar] [CrossRef] [PubMed][Green Version] - Casey, J.L.; Brown, T.L.; Colan, E.J.; Wignall, F.S.; Gerin, J.L. A genotype of hepatitis D virus that occurs in northern South America. Proc. Natl. Acad. Sci. USA
**1993**, 90, 9016–9020. [Google Scholar] [CrossRef] [PubMed][Green Version] - Jukes, T.H.; Cantor, C.R. Evolution of protein molecules. In Mammalian Protein Metabolism, 3rd ed.; Munro, H.N., Ed.; Academic Press: Cambridge, MA, USA, 1969; pp. 121–132. [Google Scholar]
- Korber, B. HIV Signature and Sequence Variation Analysis. In Computational and Evolutionary Analysis of HIV Molecular Sequences; Rodrigo, A.G., Learn, G.H., Jr., Eds.; Kluwer Academic Publishers: Dordrecht, The Netherlands, 2000; pp. 55–72. [Google Scholar]
- Miyata, T.; Yasunaga, T. Molecular evolution of mRNA: A method for estimating evolutionary rates of synonymous and amino acid substitutions from homologous nucleotide sequences and its application. J. Mol. Evol.
**1980**, 16, 23–36. [Google Scholar] [CrossRef] [PubMed] - Nei, M.; Gojobori, T. Simple methods for estimating the numbers of synonymous and nonsynonymous nucleotide substitutions. Mol. Biol. Evol.
**1986**, 3, 418–426. [Google Scholar] [CrossRef] [PubMed] - R Core Team. R: A language and environment for statistical computing. In R Foundation for Statistical; Computing: Vienna, Austria, 2018; Available online: https://www.R-project.org/ (accessed on 15 February 2021).
- Usman, Z.; Velkov, S.; Prozer, U.; Roggendorf, M.; Frishman, D.; Karimzadeh, H. HDVdb: A comprehensive hepatitis D virus database. Viruses
**2020**, 12, 538. [Google Scholar] [CrossRef] [PubMed]

**Figure 1.**Schematic diagram of the aligned HDV genome. The three red rectangles indicate the non-coding regions. The numbers colored red indicate the positions of the non-coding regions in the alignment shown in Supplementary Data S1. The blue rectangle indicates the negative strand coding region of the L-HDAg gene. The numbers colored blue indicate the positions of the coding region in the alignment. The black line indicates the region that includes the complementary sites of the coding region.

**Figure 2.**Procedure of the distance ratio analysis. (

**a**) Calculation of distances of two regions between every possible pair of aligned sequences. The symbols, d

_{a}(x, y) and d

_{b}(x, y), indicate the evolutionary distance between sequences x and y in region a and that in region b. The distances d

_{a}(x, y) and d

_{b}(x, y) correspond to one of the five distances ed

_{bp}(x, y), ed

_{nbp}(x, y), ed(x, y), ${K}_{s}^{c}\left(x,y\right),$ and ${K}_{a}^{c}\left(x,y\right)$. For simplicity of the explanation, the two regions are drawn as separate regions in an alignment. However, the regions can be overlapped. For example, when distance ratios of ${K}_{a}^{c}\left(x,y\right)$ and ${K}_{s}^{c}\left(x,y\right)$ are examined as d

_{a}(x, y) and d

_{b}(x, y), the regions a and b are the non-synonymous and synonymous sites of the same coding region. (

**b**) Calculation of distance ratios between every possible pair of aligned sequences. If d

_{b}(x, y) equals zero, the corresponding distance ratio is not included in the set of the distance ratios. (

**c**) One-sample Wilcoxon test. There are two alternative hypotheses, the median of the ratios > 1.0 and the median of the ratios < 1.0, which correspond to the right-sided and left-sided test. One of the two alternative hypotheses was adopted based on the median of the calculated distance ratios. That is, when the median was larger than 1.0 (less than 1.0), the right-sided (left-sided) one-sample Wilcoxon test was applied to the data of distance ratios.

**Figure 3.**Box plots of ten distance ratios for three genotypes. The ordinate indicates the values of the distance ratios. The symbols on the abscissa, A–J, correspond to the ratios ed

_{bp}(x,y)/ed

_{nbp}(x, y), ed

_{bp}(x,y)/ed(x, y), ed

_{nbp}(x,y)/ed(x, y), ed

_{bp}(x,y)/${K}_{s}^{c}\left(x,y\right)$, ed

_{nbp}(x,y)/${K}_{s}^{c}\left(x,y\right)$, ed(x,y)/${K}_{s}^{c}\left(x,y\right)$, ed

_{bp}(x,y)/${K}_{a}^{c}\left(x,y\right)$, ed

_{nbp}(x,y)/${K}_{a}^{c}\left(x,y\right)$, ed(x,y)/${K}_{a}^{c}\left(x,y\right)$, and ${K}_{a}^{c}\left(x,y\right)$ /${K}_{s}^{c}\left(x,y\right)$. The top and the bottom of a box indicates the upper and lower quantile of the distance ratios. The horizontal bar in the box indicates the median of the distance ratios. The top and the bottom of the line running through the box indicates the maximum and minimum of the ratios, except for the outliers. The ratios that are present outside a closed interval, $\left[\mathrm{lower}\mathrm{quantile}-1.5\times \left(\mathrm{upper}\mathrm{quantile}-\mathrm{lower}\mathrm{quantile}\right),\mathrm{upper}\mathrm{quantile}+1.5\times \left(\mathrm{upper}\mathrm{quantile}-\mathrm{lower}\mathrm{quantile}\right)\right]$ are regarded as the outliers. The circles above or below the line indicate the outliers.

**Figure 4.**Box plots of the ten distance ratios between genotypes. The ordinate indicates the values of the distance ratios. The symbols on the abscissa, A–J, correspond to ed

_{bp}(x,y)/ed

_{nbp}(x, y), ed

_{bp}(x,y)/ed(x, y), ed

_{nbp}(x,y)/ed(x, y), ed

_{bp}(x,y)/${K}_{s}^{c}\left(x,y\right)$, ed

_{nbp}(x,y)/${K}_{a}^{c}\left(x,y\right)$, ed(x,y)/${K}_{a}^{c}\left(x,y\right)$, ed

_{bp}(x,y)/${K}_{a}^{c}\left(x,y\right)$, ed

_{nbp}(x,y)/${K}_{s}^{c}\left(x,y\right)$, ed(x,y)/${K}_{s}^{c}\left(x,y\right)$, and ${K}_{a}^{c}\left(x,y\right)$/${K}_{s}^{c}\left(x,y\right)$. See legend of Figure 3 for the explanation of box plot.

Genotype I (170 Sequences) |

HQ005371, MN984413, MN984411, MN984408, MN984443, MN984429, KJ744224, MN984407, HQ005367, MN984422, MN984459, MN984449, MN984415, MN984453, MN984452, MN984455, MN984451, MN984450, MN984454, MN984427, MN984437, MN984424, MN984428, MN984457, MN984448, MN984425, MN984447, MN984421, MN984430, MN984423, MN984461, MN984414, MN984432, MN984410, MN984431, MT583796, MN984412, MN984460, MN984416, KJ744217, KJ744216, KJ744215, KJ744214, MN984435, MN984436, MN984409, MN984445, MN984419, MN984441, AB118848, MN984440, MN984438, MN984442, MN984417, MN984439, MN984420, MN984466, MN984458, MH457143, X85253, MT583812, KJ744223, KJ744220, KJ744222, KJ744221, KT722840, MN984456, KF660602, AB118849, MN984426, AY648957, AY648956, KF660600, MN984444, MN984434, AY648959, AF425644, AY648958, AF104263, MH791030, MH791028, KY463681, U81989, MG926381, MK890226, MK890225, MK890231, HQ005372, MG926380, HQ005366, MN984469, MH791029, MK890228, MK890227, MK890232, MK890235, KJ744237, MK890234, HQ005370, HQ005364, MK890230, HQ005368, HQ005365, U81988, AY633627, KJ744255, MN984418, KJ744238, MH791027, MT583805, MT583804, MH457145, KJ744243, MN984463, MN984433, KJ744257, KJ744244, EF514905, EF514907, EF514904, EF514903, EF514906, KJ744256, MN984465, KJ744228, KJ744227, KJ744234, KJ744254, KJ744248, KJ744232, KJ744253, KJ744247, MN984462, KJ744245, MN984464, KJ744218, KJ744249, KJ744231, MN984446, KJ744226, KJ744225, NC_001653, D01075, KJ744230, KJ744229, KJ744235, KJ744233, KJ744250, MK124579, M21012, HM046802, AJ307077, AJ000558, HQ005369, KJ744242, KJ744240, KJ744241, MK890224, MK890229, MG711778, MK890233, KM110794, KM110792, KM110797, KM110799, MG711717, KM110795, KM110791, KM110798, KM110790 |

Genotype II (51 sequences) |

KF660599, KF660598, MG557658, AB118846, AY261457, AF425645, AF104264, MK234591, MK234593, MK234592, MK234594, MG557659, MN984468, KM110805, MN984470, MG711777, AB118844, AB118842, AF209859, MT050453, AB118843, AY648954, AY648953, AY648952, AB118847, AY648955, AB118841, MN401236, AB118845, AB118826, AB118835, AB118834, AB118819, AB118828, AB118832, AB118840, AB118830, AB118827, AB118820, AB118825, AB118821, AB118839, AB118822, AB118818, AB118833, AB118838, AB118823, AB118836, AB118824, AB118837, AB118831 |

Genotype III (5 sequences) |

AB037947, KC590319, HF679406, HF679405, HF679404 |

Threshold | Average Ratio |
---|---|

0.4 | 0.753 |

0.5 | 0.722 |

0.6 | 0.672 |

0.7 | 0.623 |

**Table 3.**Medians and corrected p-values of ten distance ratios for three genotypes. The size of the samples of each genotype is indicated by n. The corrected p-values were calculated by the left-sided test (the alternative hypothesis is that the median is less than 1.0) or the right-sided test (the alternative hypothesis is that the median is greater than 1.0.). The asterisk indicates the p-value calculated by the right-sided test. The floating point number representation is used to express the p-values.

ed_{bp}(x,y)/ed_{nbp}(x, y) | ed_{bp}(x,y)/ed(x, y) | ed_{nbp}(x,y)/ed(x, y) | ed_{bp}(x,y)/${\mathit{K}}_{\mathit{s}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ed_{nbp}(x,y)/${\mathit{K}}_{\mathit{s}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ed(x,y)/${\mathit{K}}_{\mathit{s}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ed_{bp}(x,y)/${\mathit{K}}_{\mathit{a}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ed_{nbp}(x,y)/${\mathit{K}}_{\mathit{a}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ed(x,y)/${\mathit{K}}_{\mathit{a}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ${\mathit{K}}_{\mathit{a}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$${\mathit{K}}_{\mathit{s}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ||
---|---|---|---|---|---|---|---|---|---|---|---|

I | median | 0.46 | 0.59 | 1.3 | 0.16 | 0.34 | 0.27 | 0.52 | 1.1 | 0.89 | 0.30 |

n = 14,276 | p-value | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} * | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} * | <2.2 × 10^{−15} | <2.2 × 10^{−15} |

II | median | 0.70 | 0.71 | 1.0 | 0.30 | 0.44 | 0.43 | 0.87 | 1.2 | 1.2 | 0.35 |

n = 1263 | p-value | <2.2 × 10^{−15} | <2.2 × 10^{−15} | 8.6 × 10^{−4} * | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} * | <2.2 × 10^{−15} * | <2.2 × 10^{−15} |

III | median | 0.29 | 0.50 | 1.3 | 0.40 | 0.74 | 0.63 | 0.63 | 1.8 | 1.1 | 0.69 |

n = 9 | p-value | 1.9 × 10^{−2} | 1.9 × 10^{−2} | 1.4 × 10^{−1} * | 2.0 × 10^{−2} | 1.0 * | 1.0 * | 2.0 × 10^{−2} | 2.0 × 10^{−1} * | 6.4 × 10^{−1} * | 1.0 |

**Table 4.**Medians and corrected p-values of ten distance ratios between genotypes. The size of the samples of each genotype is indicated by n. See legend of Table 3 for the asterisk and the representation of the p-value.

ed_{bp}(x,y)/ed_{nbp}(x, y) | ed_{bp}(x,y)/ed(x, y) | ed_{nbp}(x,y)/ed(x, y) | ed_{bp}(x,y)/${\mathit{K}}_{\mathit{s}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ed_{nbp}(x,y)/${\mathit{K}}_{\mathit{s}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ed(x,y)/${\mathit{K}}_{\mathit{s}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ed_{bp}(x,y)/${\mathit{K}}_{\mathit{a}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ed_{nbp}(x,y)/${\mathit{K}}_{\mathit{a}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ed(x,y)/${\mathit{K}}_{\mathit{a}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ${\mathit{K}}_{\mathit{a}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)/{\mathit{K}}_{\mathit{s}}^{\mathit{c}}\left(\mathit{x},\mathit{y}\right)$ | ||
---|---|---|---|---|---|---|---|---|---|---|---|

I (vs.) II | median | 0.76 | 0.74 | 0.99 | 0.21 | 0.28 | 0.28 | 0.70 | 0.89 | 0.93 | 0.31 |

n = 8670 | p-value | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <1.8 × 10^{−7} * | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} |

I (vs.) III | median | 0.60 | 0.71 | 1.2 | 0.26 | 0.43 | 0.36 | 0.87 | 1.5 | 1.2 | 0.29 |

n = 850 | p-value | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} * | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} * | <2.2 × 10^{−15} * | <2.2 × 10^{−15} |

II (vs.) III | median | 0.61 | 0.72 | 1.2 | 0.32 | 0.48 | 0.43 | 0.82 | 1.3 | 1.1 | 0.38 |

n = 255 | p-value | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} * | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} | <2.2 × 10^{−15} * | <2.2 × 10^{−15} * | <2.2 × 10^{−15} |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Nagata, S.; Kiyohara, R.; Toh, H.
Constraint of Base Pairing on HDV Genome Evolution. *Viruses* **2021**, *13*, 2350.
https://doi.org/10.3390/v13122350

**AMA Style**

Nagata S, Kiyohara R, Toh H.
Constraint of Base Pairing on HDV Genome Evolution. *Viruses*. 2021; 13(12):2350.
https://doi.org/10.3390/v13122350

**Chicago/Turabian Style**

Nagata, Saki, Ryoji Kiyohara, and Hiroyuki Toh.
2021. "Constraint of Base Pairing on HDV Genome Evolution" *Viruses* 13, no. 12: 2350.
https://doi.org/10.3390/v13122350