Jakub Barylski

This paper organizes a community challenge to develop automated virus taxonomy classification tools and evaluates their performance on a combined dataset of expert-curated and metagenomic sequences.

Vclust clusters viral genomes efficiently and accurately using Lempel-Ziv parsing for average nucleotide identity, outperforming existing tools.

The paper presents a gold-standard dataset of manually curated bacterial genomes and a framework to evaluate prophage prediction algorithms, showing their performance metrics across seven tools.

Phirbo improves phage-host relationship prediction by integrating sequence similarity and bacterial relatedness, outperforming existing tools in precision and recall.

RaFAH is a machine-learning method for virus-host prediction that outperforms existing methods with an F1-score of 0.97 at the phylum level, discovering new archaeal virus lineages and auxiliary metabolic genes.