Phage-Host Prediction Using a Computational Tool Coupled with 16S rRNA Gene Amplicon Sequencing

Area of Science:

• Environmental metagenomics
• Virome analysis
• Microbial ecology

Background:

• Metagenomics reveals vast viral diversity in aquatic ecosystems, but genomic data remains largely unannotated.
• Lack of accurate phage-host information hinders phage identification and interaction studies.
• Aquatic environments like Ebrié Lagoon harbor complex phage communities.

Purpose of the Study:

• To develop and apply a viral metagenomic framework for deciphering phage community composition and diversity.
• To predict bacterial hosts for identified phages using a machine learning-based tool.
• To investigate phage-host interactions in relation to environmental factors (water vs. sediment).

Main Methods:

• Collected water and sediment samples from Ebrié Lagoon, Ivory Coast.
• Characterized bacterial communities using 16S rRNA metabarcoding.
• Analyzed virome datasets using VirSorter, VIBRANT, MetaPhinder, and Prokaryotic Virus-Host Predictor for phage identification, taxonomic annotation, and host prediction.

Main Results:

• Distinct taxonomic profiles of phages and bacteria were observed between water and sediment samples.
• Specific phage families (Microviridae in sediment; Siphoviridae, Myoviridae, Podoviridae in water) showed varied distribution.
• Predicted bacterial phyla distributions showed significant similarity to those identified by 16S rRNA metabarcoding, confirming the approach's reliability.

Conclusions:

• The developed metagenomic framework reliably identifies phage contigs and predicts their bacterial hosts.
• Phage and bacterial community structures are site-specific (water vs. sediment) within the Ebrié Lagoon.
• This approach advances the understanding of phage-host dynamics in aquatic microbial ecosystems.