Yanni Sun

SLAFIT infers viral fitness and detects structural variants from third-generation sequencing data by analyzing linkage disequilibrium and haplotype frequencies.

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.

VirProtRAG improves viral protein function annotation by integrating literature retrieval and evidence-based generation, introducing 32.53% new annotations and supporting sequence-inferred functions.

PlasRAG characterizes and retrieves plasmids using sequence-text alignment, showing that plasmids with broader host ranges and ARGs spread more extensively.

This paper introduces a method to infer fitness effects from short-read time-series genomic data while accounting for genetic linkage, validated on HIV and SARS-CoV-2 datasets.

SegFinder identifies RNA virus genome segments by analyzing co-occurrence across samples, discovering novel segments and expanding known segmented virus families.

MOSTPLAS improves plasmid host range prediction by addressing missing labels in training data through self-correction and pseudo label learning.

ViraLM is a viral language model that improves virus detection in metagenomic data, particularly for short contigs, by leveraging a genome foundation model and a curated dataset.

VirID improves RNA virus discovery and characterization from metagenomic data by reducing false positives, accurately estimating diversity, and identifying virus-host associations.

PlasGO predicts GO terms for plasmid-encoded proteins using a hierarchical model, achieving high precision in function annotation.

AnnoSINE_v2 improves SINE annotation in animal genomes with higher accuracy and computational efficiency compared to previous tools.

GDmicro improves disease status classification using gut microbiome data by combining semi-supervised learning and domain adaptation, achieving higher accuracy than existing methods.

VirStrain identifies viral strains from short reads using a greedy covering algorithm, outperforming existing tools on simulated and real data.

StrainScan improves strain-level microbiome analysis by using a tree-based k-mer indexing structure, achieving higher accuracy and resolution than existing tools.

CHEER improves taxonomic classification of viral metagenomic data by using a hierarchical deep learning model, achieving higher accuracy than existing methods.

The paper introduces VirBin, a tool for clustering viral contigs into haplotypes, showing improved sensitivity and precision over existing methods.

This paper introduces PEHaplo, a tool for de novo haplotype reconstruction in viral quasispecies, showing improved performance over existing methods.