PigViroScope: Predefined a priori protocol for a systematic mapping review to identify and catalogue pig-infecting viruses documented in the scientific literature

PigViroScope is a predefined, a priori protocol for a systematic mapping review to identify and catalogue viruses reported to infect, be detected in, be isolated from, or cause disease in members of the family Suidae, including domestic pigs, wild boar, and other wild suids. The review question was structured using the Population–Concept–Context (PCC) framework. Systematic searches will be conducted in PubMed, Scopus, and Web of Science using database-specific search strings combining population terms covering the family Suidae with virus-related terms. Because of the anticipated volume of literature, both eligibility assessment and data extraction will be conducted on titles and abstracts rather than full texts. Records will be deduplicated in R and screened using EULE (Evidence Unification through Language-model Evaluation), a locally deployed large-language-model pipeline. Sixteen prompt configurations will be evaluated using Bayesian latent class analysis, and the configuration identified as optimal will be applied for title-and-abstract screening. For included records, the review will extract bibliographic information, host type and geography, virus name and taxonomic family, detection method and sample type, study design, number of animals, and clinical signs or pathological findings where explicitly reported. The purpose of the review is to provide a comprehensive map of reported virus–Suidae associations and to describe the structure of the available evidence. Meta-analysis and formal risk-of-bias assessment are not planned. This protocol was finalised and publicly deposited prior to commencement of the formal literature search and screening process. Any subsequent amendments affecting the conduct of the review will be documented transparently in later protocol versions. This work was conducted within the European Partnership on Animal Health and Welfare and co-funded by the European Union's Horizon Europe programme under grant agreement No. 101136346 (EUPAHW). Views and opinions expressed are those of the authors only and do not necessarily reflect those of the European Union or the European Research Executive Agency.