First evidence of allexiviruses in allium plants in Ukraine and molecular characterization of their isolates Authors: Kyrylo Taher, Tetiana Shevchenko, Halyna Snihur, Oleksiy Shevchenko, Iryna Budzanivska Published: 2026 ABSTRACT Allexiviruses (family Alphaflexiviridae) are widespread pathogens of vegetatively propagated allium crops, but their occurrence has not previously been documented in Ukraine. We surveyed cultivated allium plants collected in eight Ukrainian regions (2022–2025) and screened their samples for garlic virus B (GarV-B), garlic virus C (GarV-C) and shallot virus X (ShVX) using enzyme-linked immunosorbent assay (ELISA). GarV-B, GarV-C and ShVX were detected in 39/108 (36.1%), 23/108 (21.3%) and 21/108 (19.4%) plants, respectively, with infections which were strongly host-associated: garlic (n = 63) had high frequencies of indicated viruses (GarV-B—61.9%; GarV-C—36.5%; ShVX—28.6%), whereas onion samples (n = 33) were largely negative (ShVX—3.0%; GarV-B and GarV-C—not detected). Co-occurrence analysis within garlic revealed a nested allexivirus module in which GarV-C and ShVX occurred only in GarV-B-positive plants. RT-PCR and Sanger sequencing generated 11 partial genomes representing GarV-B, GarV-C, ShVX, GarV-A and GarV-D. Maximum-likelihood phylogenies placed Ukrainian allexivirus isolates within established global diversity and indicated both European- and Asian-affiliated lineages. These findings provide the first evidence of allexiviruses in Ukrainian allium crops, and support their inclusion in plant health surveillance and planting-material certification.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-47632-1. Introduction Allium crops such as Allium sativum L., Allium cepa L. var. cepa, and Allium cepa L. var. aggregatum are of high economic importance worldwide, but their production is often constrained by viral diseases. These crops are susceptible to a range of viruses, especially those in the genera Potyvirus, Carlavirus, and Allexivirus1. Garlic is propagated vegetatively, while onion and shallot can be seed-propagated but are also frequently propagated vegetatively, which can facilitate virus persistence and dissemination through planting material. Among these, allexiviruses (family Alphaflexiviridae) are particularly notable pathogens of Allium species. The genus Allexivirus currently comprises eight recognized species infecting alliaceous plants: garlic viruses A, B, C, D, E, X, garlic mite-borne filamentous virus, and shallot virus X2. Allexiviruses are flexuous filamentous viruses with positive-sense single-stranded RNA genomes of approximately 8–9 kb, bearing a 5′ cap and a 3′ poly(A) tail. In the allium-infecting allexiviruses, the genome typically contains seven ORFs encoding a replicase, triple gene block movement proteins, a 42-kDa protein, the coat protein, and a nucleic acid-binding protein3. In addition to efficient dissemination through vegetative propagation of infected bulbs/cloves, several allexiviruses can also be transmitted by eriophyid mites, providing a mechanism for secondary spread within allium plantings4. Allexivirus infections in allium plants are typically latent or mild in symptomatology, often causing only subtle mosaic, mottling, faint chlorotic striping of leaves, or mild stunting4,5. Despite the limited visible symptoms, these viruses can have a significant economic impact on allium crops. Infected plants suffer yield losses and quality deterioration of bulbs6. For example, in a single year, infections with garlic virus A (GarV-A) and garlic virus C (GarV-C) were reported to reduce bulb weight by up to 32% and 15%, respectively6. Mixed infections, especially with potyviruses or carlaviruses, can result in even greater yield reductions7,8. Consequently, allexiviruses are regarded as important contributors to the garlic and onion viral disease complex that undermines crop productivity and bulb quality4,8. In allium crops, allexiviruses are ubiquitous. First allexivirus infecting Allium species was detected in the early 1990s in Russia9. Since then, these viruses have been reported across Asia, Europe, Africa, Oceania, and the Americas. For instance, Allexivirus species have been identified in garlic or shallot in Japan, China, and India in Asia, Greece, Italy, Spain, and Poland in Europe, as well as in the Middle East and Africa (e.g., Iran, Sudan) and the Americas (e.g., Argentina, USA)10–13. Notably, Ukraine-neighboring countries in Eastern Europe have reported the presence of allexiviruses—for example, Russia and Poland have documented multiple garlic-infecting Allexivirus species9–14. In Ukraine, however, there have been no prior records of allexivirus infections, underscoring a significant gap in the regional knowledge of allium virus distribution. Here we report the first detection and characterization of allexiviruses in garlic, onion, and shallot in Ukraine. Because allium plants are frequently affected by mixed viral infections, we also considered allexivirus occurrence in relation to other major allium viruses detected in the same sample set, namely leek yellow stripe virus (LYSV), onion yellow dwarf virus (OYDV), shallot yellow stripe virus (SYSV), garlic common latent virus (GCLV), and shallot latent virus (SLV)15,16, all of which have previously been found in Ukraine. This study provides the first evidence of allexivirus presence in allium crops in Ukraine, thus extending the current knowledge of these viruses and providing an initial step toward assessing their prevalence and impact on Ukraine’s allium production and export. The findings contribute important new data on Allium viral pathogens in the region, with implications for crop health monitoring and management in Ukrainian agriculture. Results Prevalence of allexiviruses in allium plants in Ukraine ELISA screening detected allexiviruses in the surveyed Allium material (Table 1). Overall, GarV-B, GarV-C, and ShVX were detected in 39/108 (36.1%), 23/108 (21.3%), and 21/108 (19.4%) samples, respectively. Detections were concentrated primarily in garlic (A. sativum), in which GarV-B, GarV-C, and ShVX occurred in 39/63 (61.9%), 23/63 (36.5%), and 18/63 (28.6%) samples, respectively.Table 1Serological screening of different Allium species and Ukraine’s regions for virus occurrence.SpeciesGarV-BGarV-CShVXGarlic (A. sativum) Cherkasy region1/130/130/13 Kyiv region14/1810/188/18 Odesa region7/86/83/8 Poltava region2/42/41/4 Ternopil region4/40/40/4 Vinnytsia region6/113/114/11 Zakarpattia region3/32/31/3 Zaporizhzhia region2/20/21/2Subtotal (Garlic)39/6323/6318/63Onion (A. cepa) Cherkasy region0/30/30/3 Kyiv region0/100/101/10 Odesa region0/80/80/8 Poltava region0/10/10/1 Ternopil region0/30/30/3 Vinnytsia region0/80/80/8Subtotal (Onion)0/330/331/33Leek (A. ampeloprasum)0/40/40/4Shallot (A. cepa var. aggregatum)0/30/32/3A. fistulosum0/10/10/1A. schoenoprasum0/10/10/1A. scorodoprasum0/10/10/1A. tuberosum0/10/10/1A. hollandicum0/10/10/1Total39/10823/10821/108 Table 1Serological screening of different Allium species and Ukraine’s regions for virus occurrence.SpeciesGarV-BGarV-CShVXGarlic (A. sativum) Cherkasy region1/130/130/13 Kyiv region14/1810/188/18 Odesa region7/86/83/8 Poltava region2/42/41/4 Ternopil region4/40/40/4 Vinnytsia region6/113/114/11 Zakarpattia region3/32/31/3 Zaporizhzhia region2/20/21/2Subtotal (Garlic)39/6323/6318/63Onion (A. cepa) Cherkasy region0/30/30/3 Kyiv region0/100/101/10 Odesa region0/80/80/8 Poltava region0/10/10/1 Ternopil region0/30/30/3 Vinnytsia region0/80/80/8Subtotal (Onion)0/330/331/33Leek (A. ampeloprasum)0/40/40/4Shallot (A. cepa var. aggregatum)0/30/32/3A. fistulosum0/10/10/1A. schoenoprasum0/10/10/1A. scorodoprasum0/10/10/1A. tuberosum0/10/10/1A. hollandicum0/10/10/1Total39/10823/10821/108 Serological screening of different Allium species and Ukraine’s regions for virus occurrence. In onion (A. cepa), GarV-B and GarV-C were not detected, and ShVX was found in only 1/33 (3.0%) samples. In shallot (A. cepa var. aggregatum), ShVX was detected in 2/3 samples, whereas GarV-B and GarV-C were not detected. All tested samples of leek (A. ampeloprasum), A. fistulosum, A. schoenoprasum, A. scorodoprasum, A. tuberosum, and A. hollandicum were negative for all three viruses. Virus-virus associations To place allexivirus detections in a mixed-infection context, we analyzed co-occurrence with potyvirus and carlavirus detections obtained in the same plants; prevalence results for those viruses in this sample set have been reported previously15,16. Using the complete-case dataset for the eight-virus panel (LYSV, OYDV, SYSV, GCLV, SLV, GarV-B, GarV-C, ShVX; n = 108), allexiviruses were commonly observed within multi-virus profiles, including combinations where two or more allexiviruses occurred together and/or co-occurred with potyviruses/carlaviruses. Because allexivirus detections were host-associated, pairwise virus-virus associations were evaluated primarily within garlic (A. sativum) using the complete-case dataset for the eight-virus panel (LYSV, OYDV, SYSV, GCLV, SLV, GarV-B, GarV-C, ShVX; n = 63). Two-sided Fisher’s exact tests and Pearson chi-square tests applied to 2 × 2 presence/absence tables identified a strongly structured allexivirus module in which GarV-B formed the core. GarV-C was never detected in the absence of GarV-B (23/23 GarV-C-positive samples were also GarV-B-positive; Fisher p = 4.39 × 10⁻⁷; χ2 p = 2.0 × 10⁻⁶). Likewise, ShVX was found only in GarV-B-positive plants (Fisher p = 2.89 × 10⁻5; χ2 p = 8.2 × 10⁻5), and GarV-C and ShVX co-occurred significantly (Fisher p = 1.95 × 10⁻⁶; χ2 p = 1.0 × 10⁻⁶). Using potyvirus and carlavirus detections specifically to evaluate the co-infection context of allexiviruses, GarV-C showed positive associations with LYSV (Fisher p = 0.00847; χ2 p = 0.00538) and with SLV (Fisher p = 0.00835; χ2 p = 0.00596) in garlic. In addition, LYSV and SLV co-occurred more frequently than expected by chance (Fisher p = 0.00231; χ2 p = 0.00160). Molecular confirmation and phylogenetic placement of allexivirus isolates To complement ELISA screening, ELISA-positive samples were also characterized by RT-PCR, with identical results, and Sanger sequencing of partial allexivirus genomes. In total, 11 sequences were obtained. Amplicons generated with primers targeting garlic virus B yielded four sequences that, upon sequence identification, corresponded to two GarV-B isolates (GVB-36-23, PX761687, 542 bp; GVB-43-23, PX761688, 545 bp), one GarV-A isolate (GVA-112-22, PX766271, 543 bp), and one GarV-D isolate (GVD-6-22, PX766272, 371 bp). Additional sequencing covered five GarV-C isolates (GVC-112-22, PX757648, 597 bp; GVC-12-25, PX757649, 653 bp; GVC-16-24, PX757650, 620 bp; GVC-43-23, PX757651, 569 bp; GVC-44-23, PX757652, 629 bp) and two ShVX isolates (ShVX-115-22, PX766273, 499 bp; ShVX-32-23, PX766274, 501 bp). Maximum-likelihood-based phylogeny reconstructed with representative sequences placed all Ukrainian isolates within their expected species clades and clarified their closest relationships. Two Ukrainian GarV-B sequences (GarV-B-36-23, PX761687; GarV-B-43-23, PX761688) were closely identical (99.04% nt identity), and were placed adjacent to the identical Czech isolates JX682834 and JX682834, sharing with them 98.08% and 98.28% nt identity, respectively (Fig. 1). The Ukrainian GarV-A isolate (GarV-A-112-22, PX766271) clustered closest to the Chinese sequence MN059305 (96.49% nt identity) (Fig. 2). The Ukrainian GarV-D isolate (GarV-D-6-22, PX766272) grouped with the Chinese isolates MN059369 and MN059370, showing 95.90% nucleotide identity to both (Fig. 3).Fig. 1Phylogenetic analysis of Ukrainian isolates of garlic virus B (GarV-B) using a maximum likelihood approach based on partial CP and NABP genes’ sequences. The analysis was conducted using a partitioned model with TPM2u+F+G4 for the CP partition and TN+F+G4 for the NABP partition, with 1000 bootstrap replicates. Bootstrap values > 50 are shown. Garlic virus X (GarV-X) sequence (NC_001800) served as an outgroup.Fig. 2Phylogenetic analysis of Ukrainian isolates of garlic virus A (GarV-A) using a maximum likelihood approach based on partial CP and NABP genes’ sequences. The analysis was conducted using a partitioned model with HKY+F+G4 for the CP partition and K3Pu+F+G4 for the NABP partition, with 1000 bootstrap replicates. Bootstrap values > 50 are shown. Garlic virus D (GarV-D) sequence (NC_022961) served as an outgroup.Fig. 3Phylogenetic analysis of Ukrainian isolates of garlic virus C (GarV-C) using a maximum likelihood approach based on partial CP and NABP genes’ sequences. The analysis was conducted using a partitioned model with HKY+F+I+G4 for the CP partition and TIM+F+I+G4 for the NABP partition, with 1000 bootstrap replicates. Bootstrap values > 50 are shown. Garlic virus B (GarV-B) sequence (NC_025890) served as an outgroup. Phylogenetic analysis of Ukrainian isolates of garlic virus B (GarV-B) using a maximum likelihood approach based on partial CP and NABP genes’ sequences. The analysis was conducted using a partitioned model with TPM2u+F+G4 for the CP partition and TN+F+G4 for the NABP partition, with 1000 bootstrap replicates. Bootstrap values > 50 are shown. Garlic virus X (GarV-X) sequence (NC_001800) served as an outgroup. Phylogenetic analysis of Ukrainian isolates of garlic virus A (GarV-A) using a maximum likelihood approach based on partial CP and NABP genes’ sequences. The analysis was conducted using a partitioned model with HKY+F+G4 for the CP partition and K3Pu+F+G4 for the NABP partition, with 1000 bootstrap replicates. Bootstrap values > 50 are shown. Garlic virus D (GarV-D) sequence (NC_022961) served as an outgroup. Phylogenetic analysis of Ukrainian isolates of garlic virus C (GarV-C) using a maximum likelihood approach based on partial CP and NABP genes’ sequences. The analysis was conducted using a partitioned model with HKY+F+I+G4 for the CP partition and TIM+F+I+G4 for the NABP partition, with 1000 bootstrap replicates. Bootstrap values > 50 are shown. Garlic virus B (GarV-B) sequence (NC_025890) served as an outgroup. Five Ukrainian GarV-C sequences fell within a lineage anchored by Czech isolates. GarV-C-43-23 (PX757651) and GarV-C-44-23 (PX757652) formed a strongly supported pair (99.82% nt identity) placed closest to the JX682840 Czech sequence, sharing with it 96.20% and 96.01% nt identity, respectively (and lower identity to JX682843, 93.12% and 92.93%) (Fig. 4). GarV-C-12-25 (PX757649) and GarV-C-16-24 (PX757650) formed a second Ukrainian pair (97.64% nt identity) that also showed its highest identity to JX682840 (96.56% and 95.47%), with lower identity to the Chinese sequence MN059142 (93.84% and 92.57%). The remaining Ukrainian GarV-C isolate, PX757648 (GarV-C-112-22), clustered within a phylogenetic group that included the Czech sequences JX682848, JX682851, JX682852, JX682849, JX682861, JX682846, JX682859, JX682858, JX682847, JX682857, and JX682841, as well as the Japanese sequence LC097169; within this group, it was most closely related to JX682846 (99.09% nt identity) and LC097169 (98.73% nt identity).Fig. 4Phylogenetic analysis of Ukrainian isolates of garlic virus D (GarV-D) using a maximum likelihood approach based on partial CP and NABP genes’ sequences. The analysis was conducted using a partitioned model with HKY+F+I+G4 for the CP partition and K3Pu+F+G4 for the NABP partition, with 1000 bootstrap replicates. Bootstrap values > 50 are shown. Garlic virus A (GarV-A) sequence (NC_003375) served as an outgroup. Phylogenetic analysis of Ukrainian isolates of garlic virus D (GarV-D) using a maximum likelihood approach based on partial CP and NABP genes’ sequences. The analysis was conducted using a partitioned model with HKY+F+I+G4 for the CP partition and K3Pu+F+G4 for the NABP partition, with 1000 bootstrap replicates. Bootstrap values > 50 are shown. Garlic virus A (GarV-A) sequence (NC_003375) served as an outgroup. Finally, two Ukrainian ShVX sequences clustered within established ShVX diversity represented by European and Asian isolates. ShVX-115-22 was closest to the French isolate MH389251 (89.53% nt identity). However, ShVX-32-23 showed its highest identity to the French isolates MH389250 and MH389254 (83.26% nt identity), while was more distant from the Indian isolates ON986786 and OK104171 (69.30% nt identity) (Fig. 5). Two Ukrainian ShVX sequences shared only 82.79% nucleotide identity with each other, suggesting their different origin.Fig. 5Phylogenetic analysis of Ukrainian isolates of shallot virus X (ShVX) using a maximum likelihood approach based on partial CP and NABP genes’ sequences. The tree was constructed using a TN+F+G4 model, with 1000 bootstrap replicates. Bootstrap values > 50 are shown. Garlic virus D (GarV-D) sequence (NC_0022961) served as an outgroup. Phylogenetic analysis of Ukrainian isolates of shallot virus X (ShVX) using a maximum likelihood approach based on partial CP and NABP genes’ sequences. The tree was constructed using a TN+F+G4 model, with 1000 bootstrap replicates. Bootstrap values > 50 are shown. Garlic virus D (GarV-D) sequence (NC_0022961) served as an outgroup. Discussion This study provides the first evidence that allexiviruses infect cultivated allium crops in Ukraine, extending the documented range of this group of garlic- and shallot-associated viruses in Europe2,11,14. ELISA screening revealed that GarV-B, GarV-C and ShVX were already present in surveyed material, with detections strongly structured by host: infections were frequent in garlic but rare or absent in onion. Such host skew is consistent with the epidemiology of allium viruses in vegetatively propagated garlic, where viruses can persist and accumulate across planting cycles even when symptoms are mild or unapparent1,4. Sporadic detection of ShVX in onion nevertheless indicates that allexiviruses (or at least ShVX) can occur outside garlic-dominated contexts, supporting broader host-inclusive surveillance rather than focusing solely on garlic. Allexiviruses were commonly embedded within multi-virus profiles, and the within-garlic association structure suggests non-random assembly of infections. The nested pattern in which GarV-C and ShVX were detected only in GarV-B-positive plants, together with the significant GarV-C/ShVX co-occurrence, is compatible with repeated circulation of multi-infected planting material and shared transmission opportunities, including eriophyid mite-mediated spread reported for allexiviruses4. The positive associations of GarV-C with LYSV and SLV further emphasize that allexiviruses circulate within mixed infections typical of Allium pathosystems, where co-infection can exacerbate losses relative to single infections6–8. Given that yield and bulb-quality penalties have been demonstrated for allexiviruses alone and in combination with additional viruses6–8, the structured co-occurrence observed here suggests that allexiviruses should be considered integral contributors to the virus complex affecting Ukrainian garlic, alongside the other allium viruses recently reported from the country15,16. Sequencing and phylogenetic placement confirmed multiple allexivirus species and situated Ukrainian isolates within established global diversity. Clustering of Ukrainian GarV-B and several GarV-C isolates near Czech reference lineages is consistent with regional connectivity of garlic planting material within Europe and mirrors patterns reported from other European surveys14,18. In parallel, the affinities of Ukrainian GarV-A and GarV-D to Asian isolates align with the broader global dissemination of garlic viruses documented across continents12,13, underscoring that introductions from multiple sources are plausible for clonally propagated crops. The substantial divergence between the two Ukrainian ShVX sequences supports multiple introductions and/or longer-term circulation with diversification, rather than a single recent origin, consistent with the heterogeneous ShVX diversity reported in different regions9,13. Finally, the recovery of GarV-A and GarV-D from amplicons generated with a GarV-B-targeted primer set highlights that conserved genomic regions can yield cross-species amplification; therefore, sequencing (or species-resolving assays) is essential when translating screening results into species-level inferences. Overall, the presence of multiple allexiviruses in Ukrainian allium crops – often in structured mixed infections with other economically relevant viruses—supports incorporating allexiviruses into routine monitoring and planting-material health programs. Priority next steps include expanded surveys across regions and cultivars, deeper sequence coverage to better resolve introductions and local spread, and field-based quantification of yield and quality impacts under Ukrainian (and similar) production conditions6–8. In practice, the findings reinforce the value of virus-tested planting stocks and strengthened phytosanitary oversight to limit accumulation and dissemination of allexiviruses and associated virus complexes in Ukraine’s allium sector1,6–8. Methods Plant sampling A total of 108 allium plants were analysed in Ukraine during 2022–2025, representing commercial plantings and private small-scale cultivation in eight regions (Cherkasy, Kyiv, Odesa, Poltava, Ternopil, Vinnytsia, Zakarpattia, and Zaporizhzhia). The dataset included primarily garlic (A. sativum; n = 63) and onion (A. cepa var. cepa; n = 33), with smaller numbers of shallot, leek and other Allium spp. Plants were sampled irrespective of symptom status; both plants showing virus-like symptoms (e.g., striping, mosaic, deformation) and plants without obvious symptoms were included. Serological screening (ELISA) All samples were screened serologically for allexiviruses (GarV-B, GarV-C, ShVX) using commercial kits for enzyme-linked immunosorbent assay (DSMZ, Germany) following the microplate ELISA approach17. The results were recorded using an automatic Thermo Labsystems Opsis MR microplate reader (USA) with Dynex Revelation Quicklink software at the wavelength of 405 nm. RNA extraction, RT-PCR and Sanger sequencing For molecular confirmation and phylogenetic characterization, total RNA was extracted from approximately 100 mg of leaf tissue using the Quick-RNA Plant Miniprep Kit (Zymo Research, USA) following the manufacturer’s protocol. Reverse transcription was performed with M-MuLV reverse transcriptase (NEB, USA) and a random primer mix (NEB, USA), using incubation steps of 25 °C for 5 min, 42 °C for 1 h, and 65 °C for 20 min. PCR amplification was carried out using GreenTaq PCR Master Mix (Thermo Fisher Scientific, USA). The cycling profile followed our established workflow (initial denaturation at 95 °C for 3 min, then 30 cycles of 95 °C for 30 s, annealing for 30 s, 72 °C for 45 s, and a final extension at 72 °C for 10 min), with annealing temperatures adjusted by target sequence as follows: 53 °C for GarV-B and GarV-C, and 56 °C for ShVX. Amplicons were visualized by electrophoresis in 1.5% agarose gels. Primer sets used in this study were: GarV-B-F 5′-TGACGGGCAAACAGCAGAATAA-3′ GarV-B-R 5′-ATATAGCTTAGCGGGTCCTTC-3′ and GarV-C-F 5′-TTGCTACCACAATGGTTCCTC-3′, GarV-C-R 5′-TACTGGCACGAGTTGGGAAT-3′ targeting CP+NABP region of garlic virus B and garlic virus C; ShVX-F 5′-ACCGAAATCACAGTTAACTCCTTTGG-3′ and ShVX-R 5′-TCTACGGTTGTCGATTTTGTGCGT-3′ targeting replicase gene of shallot virus X18. PCR products selected for sequencing were separated by agarose gel electrophoresis, and the target bands were excised and purified from the gel using the GeneJET Genomic DNA Purification Kit (Thermo Fisher Scientific, USA). The purified amplicons were then subjected to Sanger sequencing. Chromatograms were inspected and assembled in BioEdit v7.2.519. The obtained nucleotide sequences were deposited in the GenBank database at the National Center for Biotechnology Information (NCBI) under accession numbers provided in the Results and in the figures. Sequence alignment and phylogenetic analysis Representative reference sequences were retrieved from the GenBank for each virus and combined with Ukrainian sequences. Multiple sequence alignments were generated with MAFFT20 using the L-INS-i strategy. Maximum-likelihood phylogenetic trees were inferred in IQ-TREE21,22, with the substitution model selected using the program’s model-selection procedure (ModelFinder)23, and branch support assessed using 1,000 bootstrap replicates. For presentation, bootstrap values > 75% are shown on branches. To improve readability, densely sampled clades were collapsed where appropriate; numbers in brackets indicate the count of sequences collapsed into that node. Statistical analysis of the serological dataset Virus prevalence was summarized as proportions of virus-positive plants. In addition to allexivirus screening performed in this study, the same set of 108 allium samples had previously been examined for LYSV, OYDV, SYSV, GCLV, and SLV14,15. Mixed-infection structure and virus-virus associations were evaluated using 2 × 2 contingency tables on presence/absence calls, following standard categorical-data approaches23. Pairwise associations were tested using two-sided Fisher’s exact tests and Pearson chi-square tests, and effect size was summarized as odds ratios where applicable. Analyses were performed on complete-case subsets for the specific virus panels being compared, and results were interpreted with caution for low-frequency detections24. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1