Cross-environment genomic prediction of scale drop disease resistance in barramundi (Lates calcarifer) using laboratory challenge and farm survival data

Abstract

Scale drop disease virus (SDDV) is a major cause of mortality and economic loss in barramundi (Lates calcarifer) aquaculture across Southeast Asia, with outbreaks resulting in up to 90% mortality. While laboratory challenge models have been developed and enable accurate, standardised measurement of resistance, the extent to which these traits predict survival under natural farm outbreaks remains poorly understood, representing a key knowledge gap for implementing genomic selection. This study evaluated the genetic relationship between laboratory-derived resistance traits and survival from a commercial SDDV outbreak using two independent laboratory challenges and one farm outbreak, involving a total of 3144 fish genotyped at ∼49 k genome-wide SNPs. Resistance traits measured in the laboratory included survival time after injection, survival status, and survival 50%. Variance components were estimated using GBLUP models, and cross-environment prediction accuracies for farm survival were assessed under four training-validation scenarios. Heritability estimates for SDDV resistance ranged from h² = 0.17 to 0.44 under laboratory conditions, while under farm conditions they were substantially higher (h² = 0.73 to 0.81). Genetic correlations between laboratory and farm SDDV resistance were high within spawning batches (r<inf>g</inf> = 0.73 to 0.86), indicating low genotype-by-environment (GxE) interactions and stable genetic ranking of related individuals across environments; across batches they were moderate (r<inf>g</inf> = 0.62 to 0.75), highlighting GxE effects and the importance of relatedness. Prediction accuracy for farm survival was highest when using farm data (accuracy = 0.54), moderate when laboratory data from related populations were used (accuracy = 0.25 to 0.38), and lowest when training on unrelated batches (accuracy = 0.17 to 0.19). These results demonstrate that SDDV resistance measured in laboratory challenges captures key genetic components relevant to farm survival, while also emphasising the importance of genetic relatedness for cross-environment genomic prediction. Laboratory challenge data therefore provide a biosecure and informative source of phenotypes, enabling genomic selection to enhance disease resilience and support sustainable aquaculture breeding programmes.