Post-transcriptional processing and modification of viral RNA, including alternative splicing, polyadenylation, and methylation, play crucial roles in regulating viral gene expression, enhancing genomic stability, and increasing replication efficiency. These processes have significant implications for viral biology and antiviral therapies. In this study, using Oxford Nanopore Technology (ONT) direct RNA sequencing (DRS), we provided a comprehensive analysis of the transcriptome and epitranscriptome features of the HIV-1 B (NL4-3) subtype strain and, for the first time, characterized these features in the CRF01_AE (GX2005002) subtype strain. We identified 11 novel splicing sites among the 61 RNA isoforms in NL4-3 and defined the splicing sites for GX2005002 based on its 63 RNA isoforms. Furthermore, we identified 74 and 79 chemically modified sites in the transcripts of NL4-3 and GX2005002, respectively. Although differences in poly(A) tail length were observed between the two HIV-1 strains, no specific correlation was detected between poly(A) tail length and the number of modification sites. Additionally, three distinct N6-methyladenosine (m⁶A) modification sites were identified in both NL4-3 and GX2005002 transcripts. This study provides a detailed analysis of post-transcriptional processing modifications in HIV-1 and suggests promising avenues for future research that could potentially be applied as new therapeutic targets in HIV treatment.

Severe Fever with Thrombocytopenia Syndrome (SFTS) is caused by the SFTS virus (SFTSV) and is associated with a high mortality rate. Although previous studies have reported RNA modifications such as m6A on SFTSV RNA, an integrated analysis of native viral transcript architecture and multiple RNA modification types within infected cells remains lacking. Here, we used Oxford Nanopore direct RNA sequencing (DRS) to analyze native SFTSV RNA in infected cells, combining strand-specific alignment, isoform reconstruction through read endpoint clustering, isoform-level quantification, and signal-level modification identification using unmodified in vitro transcripts as a baseline. This approach allowed us to construct detailed maps of the L, M, and bidirectionally encoded S segments at single-molecule, isoform-level resolution. The results reveal a “length-layering” pattern in SFTSV transcription, anchored by recurrent 3′ termination hotspots: only a few full-length transcripts dominate expression, whereas multiple reproducible truncated isoforms were associated with discrete termination windows, a pattern less consistent with random degradation alone and suggestive of regulated transcript termination. At the single-nucleotide level, the modification landscape is predominantly Ψ (pseudouridine), followed by m⁵C (5-methylcytosine), with sparse m⁶A (N6-methyladenosine). Modification hotspots are co-located across isoforms at the same genomic coordinates, exhibiting segmental/strand asymmetry, with sharper peaks on (−) RNA. These patterns provide a testable framework and raise the possibility that transcript-boundary organization and site-constrained Ψ/m5C signals may be associated with variation in viral RNA output. More broadly, isoform proportions around termination hotspots and Ψ/m5C-enriched regions at conserved sites may serve as quantitative features for characterizing viral RNA organization and prioritizing targets for future functional investigation. Our single-molecule integrated map establishes a reproducible methodological framework for studying SFTSV RNA regulation and provides a resource for future work aimed at assessing how transcript boundaries and RNA modification patterns may relate to polymerase activity and virus–host interaction.

Severe fever with thrombocytopenia syndrome (SFTS) is a highly fatal tick-borne disease caused by SFTS virus (SFTSV). The Gn/Gc glycoprotein plays a critical role in viral entry, but whether its amino acid polymorphisms directly determine neurological symptoms remains unclear. We sequenced the M segment from 99 SFTS patients (31 with neurological symptoms, 68 without) in Shandong Province, China. After removing recombinant sequences, 91 non-recombinant sequences were analyzed and genotyped based on M segment phylogeny. Systematic signature screening, phylogenetic reconstruction, genotype-stratified analysis, public database validation, cross-genotype comparison, and selection pressure inference were applied to distinguish functional determinants from genotype-specific markers. M segment-based phylogeny classified the sequences into four genotypes (D, E, F, A), with D accounting for 79.3% of NS cases. Although 10 candidate sites initially distinguished NS from Non-NS, their associations disappeared upon genotype-stratified analysis; all were low-frequency variants (<15%) in public databases. Cross-genotype comparison revealed D-specific (“GERHVGIIFL”) and E/F-specific (“DGKQMRVTSF”) patterns, none of which overlapped with positively selected sites. These findings demonstrate that the apparent NS association reflected genotype confounding rather than functional determinants, identifying the 10 sites as potential genotype D-specific markers with high discriminatory potential, warranting further validation in independent datasets. This study provides a cautionary example for virus–phenotype association studies, emphasizing that genotype stratification is essential before interpreting amino acid differences as functional.