In silico prediction of potential indigenous microbial biomarkers in Penaeus vannamei identified through meta-analysis and genome-scale metabolic modelling

Abstract

Abstract Background Understanding the microbiome is crucial as it contributes to the metabolic health of the host and, upon dysbiosis, may influence disease development. With the recent surge in high-throughput sequencing technology, the availability of microbial genomic data has increased dramatically. Amplicon sequence-based analyses majorly profile microbial abundance and determine taxonomic markers. Furthermore, the availability of genome sequences for various microbial organisms has prompted the integration of genome-scale metabolic modelling that provides insights into the metabolic interactions influencing host health. However, the analysis from a single study may not be consistent, necessitating a meta-analysis. Results We conducted a meta-analysis and integrated with constraint-based metabolic modelling approach, focusing on the microbiome of pacific white shrimp Penaeus vannamei, an extensively cultured marine candidate species. Meta-analysis revealed that Acinetobacter and Alteromonas are significant indicators of "health" and "disease" specific taxonomic biomarkers, respectively. Further, we enumerated metabolic interactions among the taxonomic biomarkers by applying a constraint-based approach to the community metabolic models (4416 pairs). Under different nutrient environments, a constraint-based flux simulation identified five beneficial species: Acinetobacter spWCHA55, Acinetobacter tandoii SE63, Bifidobacterium pseudolongum 49 D6, Brevundimonas pondensis LVF1, and Lutibacter profundi LP1 mediating parasitic interactions majorly under sucrose environment in the pairwise community. The study also reports the healthy biomarkers that can co-exist and have functionally dependent relationships to maintain a healthy state in the host. Conclusions Toward this, we collected and re-analysed the amplicon sequence data of P. vannamei (encompassing 117 healthy and 142 disease datasets). By capturing the taxonomic biomarkers and modelling the metabolic interaction between them, our study provides a valuable resource, a first-of-its-kind analysis in aquaculture scenario toward a sustainable shrimp farming.