Function-based annotation and classification of uncharacterized eukaryotic viruses in the human virome

Elucidating the complex relationship between the human virome and human health depends on identifying and characterizing the functions of viral genomes, including viral genes and the regulatory elements that control their expression. Unlike coding sequences, which are typically straightforward to identify, interpreting functions of non-coding regions is much more challenging because they lack any universal grammar. This is a critical knowledge gap because non-coding regions connect viral genomes to the host translation machinery and underpin expression of viral genes that determine the host-virus relationship. The overarching goal of this proposal is to comprehensively identify and characterize non-coding elements in the eukaryotic virome that orchestrate the translation of viral genomes. We propose a combination of innovative functional and computational approaches that leverage our recently developed strategy for quantifying the translation functions of RNA libraries of tens of thousands of sequences (Direct Analysis of Ribosome Targeting, DART). DART uniquely positions us to generate the data and train the models needed to establish a grammar for interpreting the non-coding regions of viral genomes. Aim 1 will use DART to quantify the translation functions of RNA sequence across 1592 viral reference genomes covering 101/127 of known viral families that infect humans. Translation elements will be classified by their dependence on host translation factors and capacity to escape host immune defenses. Functional elements will be refined through iterative analysis of experimental and phylogenetic variants, which will then be used to train machine learning models for annotation throughout the human virome. Because RNA often functions through structures that are conserved even across divergent sequences, Aim 2 will identify RNA structural motifs with translation functions throughout the human virome. We will use chemical probing and computational strategies to identify structured elements, DART analysis of variant libraries to define structure-dependent translation functions, and structure-aware search algorithms (i.e. Infernal) to annotate these functional elements in metagenomic sequence. Finally, because the human virome includes eukaryotic viruses with non-human hosts that are themselves components of the human microbiome, Aim 3 leverages functional characterization of translation elements to build tools to predict the host (human versus commensal eukaryote) of viruses identified in metagenomic sequence. Together, our results will provide a comprehensive catalog of translation regulatory elements annotated throughout the human virome and develop powerful new predictive tools for illuminating systems of gene regulation that are central determinants of human-virus biology. Project Number: 1U01DE035620-01 | Fiscal Year: 2026 | NIH Institute/Center: National Institute of Dental and Craniofacial Research (NIDCR) | Principal Investigator: Wendy Gilbert (+1 co-PI) | Institution: YALE UNIVERSITY, NEW HAVEN, CT | Award Amount: $582,838 | Activity Code: U01 | Study Section: Special Emphasis Panel[ZRG1 IIDA-B (50)] View on NIH RePORTER: https://reporter.nih.gov/project-details/11292937