Michael B. Hall

The paper benchmarks preprocessing pipelines for Oxford Nanopore reads in bacterial genomics, finding that no single approach optimally balances variant calling and assembly performance across all metrics.

This study uses CRISPR interference and Nanopore sequencing to investigate RNA modifications in E. coli, revealing their impact on growth and protein pathways.

This study analyzes the spread of blaOXA-181-producing bacteria in Australia, finding that the IncX3 plasmid is highly stable with minimal genetic variation, suggesting that genomic similarity does not necessarily indicate recent transmission.

This paper evaluates standalone Oxford Nanopore sequencing for foodborne pathogen surveillance, showing it can achieve high accuracy with quality control tools like alpaqa.

This study uses long-read sequencing and genome graphs to analyze structural variation in Mycobacterium tuberculosis, identifying SVs linked to drug resistance and virulence.

LRGE estimates genome size from long read overlaps, outperforming k-mer-based methods and being more efficient than assembly-based approaches.

This study benchmarks variant callers on bacterial nanopore data, finding deep learning tools outperform traditional methods and Illumina in accuracy and efficiency.

A uniformly processed set of 2,440,377 bacterial and archaeal genomes with search tools and annotations, plus an AI-mined peptide set yielding ATB20, active in a mouse skin infection model.

The paper benchmarks host removal and mycobacteria classification methods in metagenomic data, finding that custom databases improve accuracy and efficiency.

This paper develops DrPRG, a tool for predicting drug resistance in Mycobacterium tuberculosis using reference graphs, showing improved sensitivity and specificity compared to existing methods.

Minos integrates variant calls from multiple tools to adjudicate discrepancies and jointly genotype bacterial cohorts, improving accuracy in detecting SNPs and indels.

This paper introduces a pan-genome graph structure and software tool, pandora, to analyze bacterial genomic variation across multiple isolates with higher SNP recovery and lower error rates than existing methods.

Chiron is a deep learning model that directly translates nanopore raw signals into DNA sequences without segmentation, achieving high accuracy and speed with minimal training data.