Flavia Villani

GNQA, a retrieval-augmented GPT question-answering system over 3000 GeneNetwork, aging and diabetes papers, cites source paragraphs; RAGAS rated answer relevance above human thumbs-up ratings (89% vs 76%).

This study quantifies protein expression in rat brains, identifies pQTLs, and links them to behavioral traits and human disorders.

This study constructs a rat pangenome from linked-read data to enhance variant discovery and genotype-phenotype mapping in recombinant inbred strains.

This paper introduces PGGB, a pipeline for building pangenome graphs that includes all variation without bias, enabling variation detection, conservation measurement, recombination event detection, and phylogenetic inference.

The paper presents an improved rat reference genome, mRatBN7.2, which enhances genomic, transcriptomic, and proteomics data mapping and identifies millions of genetic variations across rat strains.

The paper presents a draft human pangenome reference that improves variant discovery and structural variant detection compared to existing references.

The paper identifies a quantitative trait locus on chromosome 13 associated with STR expansion propensity in mice, implicating Msh3 in genome-wide mutation patterns.

The paper identifies a regulatory hotspot, meQTL.5a, that influences DNA methylation, gene expression, and aging traits in mice through genetic and age-related factors.

The paper uses linked-read whole-genome sequencing to identify DNA variants in BXD mice and maps them to phenotypes, enabling systems genetics and precision medicine.

SVJAM improves structural variant calling in linked-read data by jointly analyzing multiple samples, recovering missed variants and correcting incorrect types compared to LongRanger.

The paper introduces two C++ libraries for efficient manipulation of pangenomic graphs, showing improved performance over existing tools.