Abstract
Here we discuss three veterinary alphaherpesviruses-pseudorabies virus, equid alphaherpesvirus 1, and bovine alphaherpesvirus 1-that were instrumental in uncovering the true extent of transcriptome complexity through long-read RNA sequencing, which earlier short-read approaches could not resolve. We focus on three major transcriptomic features whose discovery and characterization relied heavily on these viral models: (i) widespread transcriptional overlaps that complicate read assignment and may drive transcriptional interference; (ii) diverse transcript isoforms arising from alternative 5' and 3' transcript termini, as well as splicing; and (iii) non-coding RNAs clustered near replication origins that illuminate potential replication-transcription interactions on a shared nuclear template. Long-read viromics in these veterinary systems has additionally served as a stringent benchmark for transcript callers and annotation pipelines, because the extreme density of overlaps and co-terminal transcript families exposes errors that often go unnoticed in typical mammalian transcriptomes. This has made veterinary herpesvirus datasets disproportionately influential in shaping best practices for full-length isoform calling, transcript end mapping, and artifact-robust cDNA library handling. We also discuss animal gammaherpesviruses as proxies for human gammaherpesviruses, allowing experimental investigation of viral programs difficult to study in human infection. Finally, we describe pseudorabies virus applications as a retrograde transneuronal tracer.