Related Experiment Video
Updated: Aug 23, 2025
08:46
Author Spotlight: Investigating Bacteriophage-Induced Immune Responses in Gnotobiotic Mice
Published on: January 26, 2024
1.8K
Host prediction for disease-associated gastrointestinal cressdnaviruses
Cormac M Kinsella1,2, Martin Deijs1,2, Christin Becker3
1Amsterdam UMC, Laboratory of Experimental Virology, Department of Medical Microbiology and Infection Prevention, University of Amsterdam, Meibergdreef 9, Amsterdam 1105 AZ, The Netherlands.
Virus Evolution
|November 3, 2022
Summary
Researchers developed a computational method to identify viral hosts in complex samples. This approach successfully predicted hosts for several gastrointestinal viruses, including those linked to human diseases like periodontitis and diarrhea.
Area of Science:
- Virology
- Computational Biology
- Microbiome Research
Background:
- Metagenomics enables virus discovery but often lacks host identification.
- This data gap hinders understanding of viral roles in human and animal diseases.
- Gastrointestinal viruses are particularly affected, limiting medical and veterinary research.
Purpose of the Study:
- To develop a computational workflow for predicting viral hosts from metagenomic data.
- To apply this workflow to identify hosts for gastrointestinal cressdnaviruses.
- To enable further research into the medical and veterinary significance of newly discovered viruses.
Main Methods:
- Developed a novel computational workflow for viral host prediction.
- Applied the workflow to 1,124 metagenomic datasets from seven gastrointestinal cressdnavirus lineages.
- Validated predictions using endogenous viral elements and case-control screening experiments.
Main Results:
- Successfully predicted hosts for four out of seven analyzed viral lineages.
- Identified Entamoeba gingivalis as the host for Redondoviridae (linked to periodontitis).
- Predicted parabasalid protists, including Dientamoeba fragilis, as hosts for Kirkoviridae.
- Linked CRESSV1 (pecoviruses) and CRESSV19 (hudisaviruses) to Blastocystis spp. and Endolimax nana, respectively, common causes of human diarrheal disease.
- Predictions for Kirkoviridae and CRESSV1 were confirmed via endogenous viral elements; Redondoviridae prediction supported by screening.
Conclusions:
- The computational workflow effectively predicts viral hosts from metagenomic data without requiring training datasets or host genome assemblies.
- This approach is adaptable to various virus lineages and significantly advances our ability to study uncharacterized viruses.
- The identified virus-host associations provide crucial insights into viral pathogenesis and the microbiome.
