Do bed bugs transmit human viruses, or do humans transmit bed bug viruses? A worldwide 2 survey of the bed bug RNA virosphere 3

,

bedbug virus 2) which extends their geographical range and the host range of Shuangao bedbug virus 1 to Cimex lectularius.We identified three novel bed bug virus sequences from a tenui-like virus (Bunyavirales), a toti-like virus (Ghabrivirales), and a luteo-like virus (Tolivirales).
Interestingly, some of the bed bug viruses branch near to insect-transmitted plant-infecting viruses, opening questions regarding the evolution of plant virus infection.When we analyzed the putative viral sequences by their host's collection location, we found unexpected patterns of geographical diversity that may reflect humans' role in bed bug dispersal.Additionally, we investigated the effect that Wolbachia, the primary bed bug endosymbiont, may have on viral abundance and found that Wolbachia infection neither promotes nor inhibits viral infection.
Finally, our results provide no evidence that bed bugs transmit any known human pathogenic viruses.
genome), their study highlights the role that bioinformatic surveillance could play in detecting human pathogen transmission in bed bugs.However, bed bug specific viruses were either not detected or not reported in their study (Ling et al., 2020).
In this study, we conducted a worldwide survey of the bed bug RNA virosphere.Our aims were to search for known human viral pathogens and novel bed bug viruses that could be of interest to human health or biocontrol.We collected bed bugs (Cimex lectularius and Cimex hemipterus) from 8 distinct locations around the world and sequenced RNA libraries from 22 individuals.We assembled bed bug metatranscriptomes, and conducted phylogenetic analyses on the virus sequences that we detected.We also assessed viral diversity between bed bug species and geographic location.Additionally, we investigated whether there is a correlation between Wolbachia (a bed bug endosymbiont known to have an antiviral effect when infecting other insect taxa) reads and bed bug virus reads in each sample (Cogni et al., 2021;Hussain et al., 2023;Lindsey et al., 2018;Teixeira et al., 2008;Terradas & McGraw, 2017).

Collection and extraction:
The samples used in this study were obtained as part of a large international collection of bed bugs provided by numerous pest control companies and researches.We included Cimex lectularius samples from Czechia (n=3), France (n=1), the UK (n=3), Rome-Italy (n=3), Assisi-Italy (n=3), Ohio-USA (n=3), the Harlan lab strain of Cimex lectularius (initially collected in Fort Dix, New Jersey, USA and maintained at the King lab at Mississippi State University) (n=3), and Cimex hemipterus samples from Madagascar (n=3).Each bug was washed in 95% ethanol and total RNA was isolated using a standard TRIzol protocol (Invitrogen Waltham, MA) and further purified using NEB's Monarch RNA cleanup kit was used (New England Biolabs Ipswich, MA).

Library Prep and Sequencing:
We checked total RNA quality using Nanodrop quantitation and agarose gel electrophoresis, and we assessed RNA integrity with an Agilent 2100 bioanalyzer.Libraries were prepared using a strand specific library prep with ribosomal RNA depletion.Sequencing was conducted on the Illumina NovaSeq 6000 instrument for 150 base pair paired-end reads, resulting in approximately 100-135 million reads per sample.The quality of the reads was inspected using FastQC v0.11.5 (Andrews, 2010) and reads were quality trimmed using trimmomatic v. 0.39 (Bolger et al., 2014).
When putative multipartite virus sequences were present, we used a viral co-occurrence detection method we previously described to identify the other genomic segments (Walt et al., 2023).Briefly, this method calculates two metrics based off sample co-occurrence.The first metric, Vco, determines the frequency at which a transcript occurs in a sample with a given viral conserved sequence (e.g., transcripts with confirmed RdRp domains detected in BLAST analysis).The second metric, Tco, determines if the transcripts found together with a viral conserved sequence occur in other samples without the viral conserved sequence.We used the thresholds Vco=0.75 and Tco=0.5 to determine candidate viral genomic segments.After running the co-occurrence analysis, we only kept candidate sequences with an ORF size > 500 nt.We further inspect these transcripts for conserved protein families and domains using NCBI's CDD and InterProScan.

Phylogeographic Analysis
We retrieved all transcripts of putative viral origin from each individual bed bug assembly using Diamond BLASTx and CD-hit output.We predicted coding sequences using EMBOSS's getorf tool using the -find 2 option.We only used coding sequences with complete RdRp domains for phylogenetic analysis and duplicate sequences within samples were discarded.We conducted phylogenetic analysis in the same way as section 2.4, except that nucleotide sequences were used instead of amino acid.We selected closely related taxa from the analysis in 2.4 as outgroups for phylogeographic analysis.We calculated evolutionary distance as p-distance using MEGA v.11.0.13 (Tamura et al., 2021).We did not conduct phylogeographic analysis on Shaungao Bedbug virus 2 because it was only detected in one location in our study.

Correlation of Wolbachia and Viral Read Abundance
We mapped the trimmed read datasets to all the viral genomes detected in this study and the Wolbachia endosymbiont of Cimex lectularius genome (GCF_000829315.1)using HISAT2 v.2.2.1 (Kim et al., 2019).We used the summary file output to obtain the percent of reads that mapped to the virus genomes or the Wolbachia genome.To correlate the Wolbachia and virus read abundances, we conducted a simple linear regression analysis in R v.4.2.2.(R Core Team, 2020), using the percentage of reads that mapped to the viral genomes versus the percentage of reads mapped to the Wolbachia genome.

Read Mapping, Assembly, and Transcript Annotation
To survey the bed bug virome, we sequenced the total RNA (ribosomal and small RNA depleted) from 22 individual bed bugs collected from the United Kingdom, France, Chechia, Italy, Madagascar, and the USA.To enrich for virus sequences in our bed bug samples, we filtered out reads that mapped to genomes of organisms that could be represented in our dataset.We classified an average of 85% of reads from each sample as bed bug (65.2% C. lectularius, 19.8% C. hemipterus), Wolbachia (2.08%), or human reads (0.07%).Those reads were not used for metatranscriptome assembly (Figure 2).We assembled the remaining reads and aligned the assemblies to NCBI's nr protein database using diamond BLASTx, which returned multiple significant hits to putative bed bug virus sequences.We detected sequences from two known negative-sense single stranded RNA (-ssRNA) viruses associated with C. hemipterus: Shuangao bed bug virus 1 (Sbbv1) and Shuangao bed bug virus 2 (Sbbv2), but we did not detect the other two known bed bug viruses in our assemblies (C.X. Li et al., 2015;Zhang et al., 2020).In addition, we detected 3 novel viral sequences.These sequences all belong to the realm Riboviria and encode for 1.) a -ssRNA genome, 2.) a positive-sense single stranded RNA (+ssRNA) genome, and 3.) a double stranded RNA (dsRNA) genome.Strikingly, all putative viral sequences detected in this study are present intercontinentally, but none were detected across all samples.

Shuangao Bedbug Virus 1
Sbbv1 is described as an unclassified rhabdovirus (-ssRNA) (NCBI:txid1608071) in the NCBI database.Our phylogenetic analysis groups Sbbv1 within the Bunyavirales order branching sister to the Hantaviridae, which is consistent with previous studies (C.X. Li et al., 2015) (Figure 3).We detected Sbbv1 sequences in C. hemipterus samples from Madagascar, and C. lectularius samples from France and Czechia.Our results extend Sbbv1's geographical range from China (where it was first detected) to Madagascar, Czechia, and France, and its host range to C. lectularius.We detected both known segments of Sbbv1: a ~7000 nt segment (L) containing the RdRp domain, and a ~3000 nt (M) segment encoding for the glycoprotein precursor.The complete M segment was detected in Czechia sample 3, France, and all three Madagascar samples, but it was only partially complete in Czechia Sample 2.

Shuangao Bedbug Virus 2
Sbbv2 is described as an unclassified -ssRNA virus (NCBI:txid1608072) with a 10,925 bp monopartite genome.Our analyses show that it groups with the Rhabdoviridae family (Figure 4), consistent with the findings of Li et al. (2015).We detected Sbbv2 sequences in C. hemipterus samples from Madagascar samples 1 and 3.The genomes we assembled of this virus were around 3 kb longer than those from Li et al. (2015), ranging from 12,802-13,477 bp.
Our results complete the genome of Sbbv2, as the extra 3 kb contains ORFs at the beginning of the genome: one encoding for a rhabdovirus nucleoprotein (CDD E-value = 1.77e -06 ), and the other encoding for what we suspect is the rhabdovirus phosphoprotein (Supplementary Figure 1).Although the latter ORF did not have any significant BLAST hits or conserved domains, we assume that it is the phosphoprotein based off synteny to other rhabdovirus genomes which generally encode for five proteins in the following order: nucleoprotein, phosphoprotein, matrix protein, glycoprotein, and large protein (Walker et al., 2022).Our study also extends the geographic range of this virus, as it was previously only detected in bed bugs from China (Li et al. 2015).

Tenuiviridae
We detected sequences that had significant BLAST hits to viruses in the family Phenuiviridae (-ssRNA) in C. lectularius samples from Czechia, France, Rome-Italy, and Assisi-Italy, and C. hemipterus samples in Madagascar.These viruses are typically multisegmented, and from our BLASTx-based analysis, we detected a ~9000 bp putative L segment.We found complete L segments in Czechia sample 3, Madagascar sample 2, Rome-Italy sample 1, and Assisi-Italy samples 1, 2, and 3. We found partial L segments in Czechia sample 2, France, and Rome-Italy sample 3. Our phylogenetic analysis groups these sequences in the family Phenuiviridae, sister to a clade including the tenuiviruses, which are a genus of hemipterantransmitted plant infecting viruses (Figure 3).We refer to this putative virus as Cimex tenui-like virus 1.
Because bunyaviruses are typically multisegmented, we used an approach previously described in Walt et al. (2023) to identify candidate genomic segments based off sample cooccurrence.We found a single transcript that met our requirements with a Vco= 0.86 and a Tco = 0.75.The transcript is 2,851 nt long and encodes for a 902 amino acid protein.Based on transcript length in comparison to Cimex tenui-like virus 1's closest BLAST hit (Solenopsis invicta virus 14: M segment length=2,705 bp), we propose this may be the M segment, which encodes a glycoprotein.InterProScan predicted two transmembrane domains in the amino acid sequence of this transcript, which is similar to the glycoprotein of Solenopsis invicta virus 14 as it also has two transmembrane domains.These results need further confirmation, as this transcript had no significant BLAST alignments to other known sequences, no predicted conserved domains in the translated protein, and no similar transcripts were found in the Italy samples, which also had Cimex tenui-like virus 1 sequences in them.

Luteoviridae
We detected a ~2800 bp luteo-like virus 1 sequence (+ssRNA) in the Harlan lab strain, Czechia, and France samples, all of which were C. lectularius.We found complete transcripts in Czechia samples 1,2, and 3, France, and Harlan Strain-USA samples 1 and 2, and found a partial transcript in Harlan Strain-USA sample 3. Our phylogenetic analysis groups these viruses within a clade of unclassified Luteoviridae, sharing a common ancestor with 4 viruses detected in mosquitoes, and 1 virus detected in an anal swab from a bird (Figure 5).This clade branches sister to a group of three viruses, Miscanthus yellow fleck virus, Rabbit luteovirus, and Arracacha latent virus E. Rabbit luteovirus was discovered in a rabbit but was assumed to have come from its diet (Tsoleridis et al., 2019), and the other two viruses are plant viruses that are likely transmitted by aphids, another hemipteran insect (Bolus et al., 2020).We refer to this putative virus as Cimex luteo-like virus 1.

Totiviridae
We detected a toti-like virus (dsRNA) sequence in C. lectularius samples from five locations: the Harlan lab strain-USA, Czechia, France, Assisi-Italy, and the UK.The complete genome is around 7.8 kb and encodes for two large ORFs.One ORF (~3880 nt) has significant BLASTx hits to other toti-like virus RdRps, and the other (~1860 nt) has significant BLAST hits to other toti-like virus proline-alanine rich proteins (Spear et al., 2010).We recovered whole or nearly complete genomes for the UK (all samples), Czechia (complete for samples 1 and 3, partial for sample 2), France, and all three Assisi-Italy samples.Phylogenetically, this virus shares a most recent common ancestor with an unclassified totivirus isolated from a flea.The virus forms a clade with other Hemiptera and Thysanoptera associated toti-like viruses along with two viruses isolated from plants (Figure 6).We hereon refer to this putative virus as Cimex toti-like virus 1.

Phylogeography of Bed Bug Viruses
All the virus sequences that we detected were distributed across geographically distant locations, so we investigated viral diversity between localities.We hypothesized that viral evolutionary relationships would reflect the differences in host species, and that within a host species, phylogenetic relationships would be reflective of geographic distance (Ballinger et al., 2022;Longdon et al., 2014).Specifically, we would expect C. lectularius and C. hemipterus to harbor closely related but distinct viral populations, and within C. lectularius, we would expect samples from Europe to be distinct from those rom North America.We only used sequences with complete RdRp domains and only included the unique sequences within individuals for our phylogeographic analysis.

Cimex tenui-like virus 1
We found Cimex tenui-like virus 1 sequences in both C. lectularius, and C. hemipterus.In the case of C. lectularius, sequences from this virus were found in one individual from Rome-Italy, all three individuals from Assisi-Italy, one individual from Madagascar, the individual from France, and one individual from Czechia.Interestingly, the Italian samples form a distinct clade from the rest of the world (Figure 7A).When we grouped the samples by clade: Italy, world, and outgroup and assessed evolutionary distance between the groups, the mean p-distance between the Italian clade and the rest of the world clade was 22.4%, while the "within group" mean p-distance was 3.0% and 1.1% for the Italian clade and the rest of the world, respectively (Table 1).Interestingly, the "within group" distance is three times higher in the samples that come from the same country (Italy) than those collected in different countries (France, Chechia, and Madagascar).
Because the samples collected in Italy had a higher "within group" mean than the samples from the rest of the world, we computed pairwise comparisons of p-distance for the samples used in the phylogeny in Figure 7A (Table 2).We found that one sequence of Cimex tenui-like virus 1 (Rome-Italy 1) is the driver of the viral genetic diversity within the group of samples collected in Italy, since the Assisi-Italy samples are identical, but the Rome-Italy 1 sample is 7.5% different from all the Assisi-Italy samples at the nucleotide level (Table 2).
Furthermore, the France and Czechia 3 Cimex tenui-like virus 1 sequences are nearly identical, while the Madagascar 2 Cimex tenui-like virus 1 sequence is only 1.7% different from both the France and the Czechia samples (Table 2), even though the Madagascar samples are C. hemipterus, while the rest of the samples are C. lectularius.

Cimex toti-like virus 1
We only found Cimex toti-like virus 1 sequences in C. lectularius samples.We found them in two of the three Czechia individuals, the France individual, all Assisi-Italy individuals, and all UK individuals.Once again, the samples collected from Italy form a distinct clade from the rest of the world (Figure 7B).We grouped the samples by the Italy clade, the rest of the world clade and the outgroup and found that there was little difference between the Italy viruses ("within group" mean p-distance = 0.3%), while rest of the world clade were more genetically distant ("within group" mean p-distance = 10%) (Table 1).The "between group" mean distance between the virus sequences detected in Italy and the virus sequences detected from the rest of the world is 16.4%.Within the rest of the world clade, the viruses from Czechia form one group and the France and UK viruses form another group, following expected geographic patterns.

Cimex luteo-like virus 1
We only found Cimex luteo-like virus 1 sequences in C. lectularius, and we detected them in all three Czechia samples, two of the Harlan Strain-USA samples, and the France individual.The tree has low resolution other than one branch containing the viral sequences detected in the Harlan Strain-USA bugs (Figure 7C).All the Cimex luteo-like virus 1 RdRpencoding transcripts have a mean p-distance of 4%.

Shuangao Bedbug Virus 1
We detected Sbbv1 sequences in all three C. hemipterus individuals collected from Madagascar, and one C. lectularius individual from Czechia.The Madagascar samples and the previously described Sbbv1 sample (detected in China) group together with 100% support (Figure 7D).This reflects host taxonomy, as the Sbbv1 sample that was first detected in China was also detected in C. hemipterus.Because the bed bugs collected in Czechia were C. lectularius, our study expands the host range of Sbbv1 to C. lectularius.Overall, these viruses are very similar to each other with an average p-distance of 2% in the RdRp-encoding transcripts.

Influence of Wolbachia on Viral Abundance
Many hypotheses have been proposed to understand why bed bugs have never been linked to pathogen transmission, including one involving their Wolbachia endosymbiont as a potential factor (Pietri, 2020).Wolbachia has formed a nutritional symbiosis with bed bugs, as it provides them with B-vitamins (Hosokawa et al., 2010).Other studies have shown that Wolbachia colonization can confer resistance to viral infection in other insects such as Drosophila, mosquitoes, and even hemipteran insects (Cogni et al., 2021;Gong et al., 2020;Lindsey et al., 2018;Teixeira et al., 2008).A study investigating the influence of bed bug Wolbachia on feline calicivirus titer has been conducted, but there was no evidence that the virus ever replicated inside of the bed bugs (Fisher et al., 2019).To investigate the effects that Wolbachia may have on viral abundance, we mapped all reads to the viral genomes detected in this study and the Wolbachia endosymbiont of Cimex lectularius genome.Table 3 shows the percentage of reads that mapped to the virus and Wolbachia genomes from each sample.We used these values to investigate the influence of Wolbachia reads on viral reads and we found no correlation (y = 0.41 -0.042x, R = -0.14, p = 0.53) between percent Wolbachia reads and percent virus reads in a sample (Figure 8).

Detection and Phylogenetics of Bed Bug viruses
Bed bugs are a worldwide urban pest, that have undergone population resurgence for the last 20-30 years.Although their capacity to transmit human disease remains unknown, interest in their vector competence is high because of the increasing frequency in outbreaks (Doggett & Lee, 2023).We did not detect any known human viruses, but our study supports that metatranscriptomic surveillance is a useful technique to detect what known or emerging pathogens bed bugs could potentially transmit.We detected two previously known virus sequences associated with C. hemipterus, and 3 novel putative bed bug viruses.
The two previously detected bed bug viruses, Sbbv1 and Sbbv2 had been found in China associated with the tropical bed bug, C. hemipterus (C.X. Li et al., 2015).Our phylogeny groups Sbbv1 with an insect virus sister to the Hantaviridae (Figure 3).This agrees with the findings of Käfer et al. (2019) and supports the hypothesis that the Hantaviridae may have originated from arthropod viruses, and subsequently shifted to infecting vertebrate hosts (Marklewitz et al., 2015).Although Sbbv1 shares a common ancestor with the hantaviruses, it is unknown whether it is of concern to humans.We extend Sbbv1's geographical range from China to Czechia, Madagascar, and France, and extend its host range to C. lectularius, as it had previously only been detected in the tropical bed bug, C. hemipterus (C.X. Li et al., 2015).
Sbbv2 provides an interesting insight to plant-infecting rhabdovirus evolution, as many economically important plant rhabdoviruses are transmitted by hemipteran insects (Whitfield et al., 2018).Our phylogeny agrees with Longdon et al. (2015), as Sbbv2 groups with insectspecific clade of rhabdoviruses that shares a common ancestor with the cytorhabdoviruses and the nucleorhabdoviruses (Figure 4).This supports the hypothesis that these viruses infected hemipteran insects before they infected plants (Longdon et al., 2015).It is interesting to note that Sbbv2 has persisted in the bed bug lineage despite shifts in feeding strategy from plants to other insects, to obligate blood feeders (Johnson et al., 2018).Our bunyavirus phylogeny (Figure 3) also suggests that the tenuiviruses, which are important insect-transmitted plant viruses, infected their insect hosts before evolving the ability to infect plants.This is indicated by Cimex tenui-like virus 1 branching sister to the tenuiviruses, and another insect virus, Wuhan horsefly virus, branching with this group (Figure 3).
Totiviruses are dsRNA viruses that typically infect fungi, but there is a growing number of toti-like viruses detected in arthropod and vertebrate metatranscriptomic studies (Tighe et al., 2022).According to our phylogenetic analysis, Cimex toti-like virus 1 falls within a clade of arthropod and plant infecting viruses, along with a toti-like virus 1 detected in an anal swab from a bird (GenBank: QKN88741.1)(Figure 6).Interestingly Cimex toti-like virus 1 shares a most recent common ancestor with a virus detected in fleas, which are also obligate blood feeders (Harvey et al., 2019).This supports the hypothesis that similarities in ecological niche could be more correlative of viral similarity than taxonomic relatedness (C.X. Li et al., 2015).
Most other viruses in this clade are Hemiptera or Thysanoptera-associated (a sister group to the Hemiptera) viruses.

Phylogeography of bed bug viruses
Although our study design limited an extensive phylogeographic analysis, we found unprecedented patterns of viral diversity.First, we found that bedbug viruses detected in this study are not geographically restricted and can infect more than one host species.We detected sequences from four of the five viruses in this study intercontinentally and sequences from two out of five viruses were found in both C. lectularius and C. hemipterus.(Figure 7).Second, if we found viruses that were present in both bed bug species, we expected these viruses to form distinct clades reflecting bed bug taxonomy, as differences in host receptors between species would add selective pressures to viral infection (Longdon et al., 2014).Our results did not match that expectation, as the Cimex tenui-like virus 1 detected in the C. hemipterus (Madagascar) samples grouped with the Cimex tenui-like virus 1 sequences from C. lectularius (France and Czechia), while other Cimex tenui-like virus 1 sequences detected in C. lectularius formed their own distinct clade (Rome-Italy and Assisi-Italy) (Figure 7A).Third, we expected that viruses from similar geographic location would form distinct phylogenetic groups.This trend was generally followed, but strikingly, in every case where viral sequences were detected in samples from Italy, the sequences form their own distinct clades separated from the rest of Europe (Figure 7A&B).Furthermore, even though the Cimex tenui-like virus 1 sequences from Italy group together phylogenetically, there is higher mean evolutionary distance within these samples than within the Cimex tenui-like virus 1 sequences from the rest of the world, which include samples from France, Czechia and Madagascar, and include two different host species (Tables 1 & 2).Previous studies of bed bug phylogeography have found low genetic diversity within bed bug infestation sites, but high genetic diversity between infestation sites even of relatively close proximity, which could be due to their dependence on humans for dispersal (Fountain et al., 2014;Saenz et al., 2012).Bed bug dispersal by humans could also explain the unexpected patterns of bed bug virus phylogeography, as the distinct groups of bed bug viruses could be explained by Italy's popularity as a travel destination, with Rome being a hotspot for tourism, and Assisi being a frequent site of pilgrimage.Along with this, the similarity between viruses detected in Madagascar and Europe could reflect travel between these two places.

Wolbachia influence on viral abundance
As an additional exploration of our dataset, we investigated if the amount of Wolbachia reads was correlated with viral abundance.It has been hypothesized that Wolbachia could have a protective effect against viral infection in bed bugs, similar to what has been seen in mosquitoes and Drosophila (Cogni et al., 2021;Fisher et al., 2019;Hussain et al., 2023;Lindsey et al., 2018;Teixeira et al., 2008;Terradas & McGraw, 2017).We used percentage of reads mapped to Wolbachia and percent mapped to the viral genomes detected in this study as a proxy of abundance.We found that there was no correlation between Wolbachia and virus abundance when a potential outlier sample was present (Figure 8).These results indicate that unlike Diptera-associated Wolbachia, bed bug Wolbachia may not confer viral resistance.
Although our experimental design was not ideal to test Wolbachia's influence on bed bug virus fitness, these results provide a preliminary look into how Wolbachia may affect viruses that replicate inside of bed bugs.

Conclusions
Our study opens interesting questions about the bed bug virosphere but does not provide evidence that bed bugs transmit human viruses.On the contrary, humans may drive bed bug virus diversity by facilitating dispersal and local extinction of host populations (Fountain et al., 2014).Future studies should assess the pathogenicity and transmission routes of these viruses to have a more comprehensive understanding of their potential in biocontrol or as emerging diseases.Along with this, a more comprehensive sampling strategy and 402 phylogeographic analysis could shed light on the interesting patterns of bed bug virus dispersal.403

Figure 1 :
Figure 1: Locations of sample collection sites.Samples were collected at nine distinct sites

Figure 2 :
Figure 2: Percentage of reads mapped to the bed bug genomes, human genome, and

Figure 3 :
Figure 3: Phylogeny of the Bunyavirales including viruses detected in this study.Viruses found

Figure 5 :
Figure 5: Phylogeny of the Tolivirales, including the ssRNA(+) virus discovered in this study.

Figure 6 :
Figure 6: Phylogeny of the Ghrabrivirales, including the dsRNA virus discovered in this study.

Figure 7 :
Figure 7: Phlyogeographic analysis of the putative bed bug viruses detected in this study.Only

Figure 8 :
Figure 8: Correlation of the percentage of reads mapped to Wolbachia endosymbiont of

Table 1 : Mean evolutionary distance between distinct clades of viruses from phylogeographic analysis.
Only Cimex tenui-like virus 1 and Cimex toti-like virus 1 are shown as they had highly supported trees containing many viral genomes.Distinct phylogenetic clades were formed from the Italy samples, the rest of the world, and the outgroups.