Fig 1.
Seasonal outbreaks patterns and detection rate of RSV in Seoul from 2010 to 2019.
Seasonal outbreaks patterns and detection rate of RSV in patients who visited Korea University Guro Hospital between January 2010 and December 2019. The detection rate of RSV-A and RSV-B was calculated as the percentage of positive cases detected among the total number of tests performed using either the multiplex RT-PCR (Seeplex® RV15 One-Step ACE Detection assay or Anyplex™ II 16 detection kit) or culture method. The x-axis of the graph represents the months of the year from January to December, while the y-axis represents the detection rate of RSV as a percentage. The orange line represents the detection rate of RSV-A, and the blue line represents the detection rate of RSV-B.
Fig 2.
Neighbor-joining (NJ) phylogenetic tree of Seoul RSV-A and RSV-B strains.
Phylogenetic trees based on the analysis of partial G gene sequence of RSV-positive samples from 2010–2019 in Seoul, (A) RSV-A and (B) RSV-B. Since the calculated average pairwise Jukes-Cantor distance was 0.08 and 0.06 for RSV-A and RSV-B, respectively, the data was suitable for building the NJ tree. The NJ trees were created using maximum composition likelihood and 1,000 bootstrap replicates by Mega X. Only bootstrap values ≥70% are shown for each tree. Strains are labeled with taxon markers of different colors and shapes depending on the collected year, as described in the table on the left-upper side of each NJ tree. Reference sequences are expressed in the following format; GenBank accession number /year/country/genotype. The prototype reference sequence for BA genotype (AY33364.1) was shaded in blue (B).
Fig 3.
Maximum likelihood (ML) phylogenetic tree of Seoul RSV-A and RSV-B strains.
ML phylogenetic trees based on the analysis of partial G gene sequence of RSV-positive samples from 2010–2019 in Seoul, (A) RSV-A and (B) RSV-B. The ML trees were created using the most suitable nucleotide substitution model (TN93 + G) with 1,000 bootstrap replicates by Mega X. Only bootstrap values ≥70% are shown for each tree. Strains are labeled with taxon markers of different colors and shapes depending on the collected year, as described in the table on the left-upper side of each ML tree. Reference sequences are expressed in the following format; GenBank accession number/year/country/genotype. The prototype reference sequence for BA genotype (AY33364.1) was shaded in blue (B).
Fig 4.
Maximum clade credibility (MCC) phylogenetic tree from Bayesian analysis of Seoul RSV-A and RSV-B strains.
Time-scaled MCC trees based on the analysis of partial G gene sequence of RSV-positive samples from 2010–2019 in Seoul, (A) RSV-A and (B) RSV-B BA. (A) The RSV-A ON1 and NA1 genotypes were shaded in green and yellow boxes, respectively. The clusters of BA genotypes were shaded in mid-grey, light-grey, light-blue, and deep-blue boxes in BA9, BA11, BA-SE1, and BA-SE, respectively. The sequences showing discrepancies (BA-DIS) compared with the NJ and ML trees were marked with black asterisks in BA-SE1.
Table 1.
Intergenotypic p-distances in the partial G gene calculated between Seoul RSV-B strains.
Fig 5.
Deduced amino acid sequences in the 2nd HVR of RSV G gene of Seoul RSV-A ON1 and RSV-B BA genotype.
Amino acid sequences alignment of the partial G gene of RSV, (A) RSV-A ON1 and (B) RSV-B BA genotype. Sequences are expressed as RSV subtype (A or B)/city (Seoul, SEL)/sample number/collected year. Alignments are shown, and residues are numbered relative to sequences of (A) prototype ON1 strain ON67-1210A (GenBank accession no. JN257693) and (B) prototype strain BA strain BA4128/99B (GenBank accession no. AY333364). The two copies of 23 amino acid regions in the RSV-A group and 20 amino acid regions in the RSV-B group are framed in red and blue, respectively. Identical residues were indicated by dots. To enable convenient visualization of shared mutations, the RSV-A strains were arranged in the same order as depicted in the maximum likelihood (ML) tree. On the other hand, the RSV-B strains were listed according to their clusters in their trees (BA9, BA11, BA-DIS, BA-SE, and BA-SE1).
Fig 6.
Shannon entropy plot combined with the frequency of amino acid variations of the 2nd HVR of G gene in (A) RSV-A ON1 and (B) RSV-B BA genotype. Entropy values of each amino acid are calculated using BioEdit. Amino acid positions with the entropy value >0.2 are considered variable and denoted with a label. Among them, the amino acid positions with variation frequency > 5.0% and the entropy value 0.2 are highlighted in bold letters for easy reference. Duplicated regions in ON1 and BA are framed in black rectangles.
Table 2.
Details of amino acid variations in the 2nd HVR of RSV G gene of Seoul RSV-A ON1 and RSV-B BA genotype.
Fig 7.
Proportion of amino acid variations in five Seoul RSV-B strains.
The variations in amino acid sequences observed within the 2nd HVR region of G gene of the Seoul RSV-B G gene were visually represented using different colors corresponding to the specific variations present, (A) BA9, (B) BA11, (C) BA-DIS, (D) BA-SE, and (E) BA-SE1. Unique variations that were not present (or hardly present) in other strains were highlighted in bold letters for each strain; (A) E305D in BA9, (C) S267P in BA-DIS, (D) E261K/L272P in BA-SE, and (E) T290I in BA-SE1.
Table 3.
Positive and negative selection sites in the partial G gene of RSV-A ON1 and RSV-B BA genotype using SLAC, FEL and MEME with a p-value threshold of 0.05 and FUBAR with a posterior probability of 0.9.
Fig 8.
Bayesian skyline plot analysis revealing the temporal changes in effective population size of respiratory syncytial virus (RSV).
Population dynamics of the G gene for (A) RSV-A and (B) RSV-B using Bayesian skyline plot analyses. The y-axis and x-axis represent the effective population size and the time in years, respectively. The estimated mean of the effective population size is represented by the bold line, while the colored margin indicates the 95% highest posterior density (HPD) interval.