Ben J. Woodcroft

This paper shows that adding noise to gene content data improves machine learning predictions of bacterial phenotypes across evolutionary time, concluding the last bacterial common ancestor was anaerobic, double-membraned, and moderately thermophilic.

The paper analyzes mobile genetic elements in thawing permafrost soils to assess their impact on microbial diversity and function using meta-omic data.

The paper introduces SMF, an algorithm to estimate bacterial and archaeal DNA prevalence in metagenomes, revealing biome-specific patterns and highlighting issues with human host DNA in public repositories.

Microflora Danica compiles 10,686 shotgun metagenomes and 449 rRNA datasets to create an atlas of Danish environmental microbiomes, revealing hidden homogeneity and novel nitrifiers.

Bin Chicken improves genome recovery from metagenomes by coassembling targeted samples, recovering 77,562 microbial genomes including many novel species.

CompareM2 is a pipeline that compares microbial genomes to generate reports on quality, annotation, taxonomy, function, and phylogeny.

SingleM and Sandpiper improve microbial taxonomic profiling in metagenomic data by detecting novel species and showing most environments are dominated by unrepresented taxa.

The paper examines microbial responses to warming in thawing permafrost peatlands over seven years, finding habitat-specific stability despite environmental changes.

The paper investigates denitrifying protist endosymbionts in wastewater treatment plants, finding their presence in 20-50% of facilities and highlighting temporal fluctuations and incomplete denitrification pathways.

The paper quantifies eukaryotic DNA in metagenomes and shows it biases microbial trait estimation, finding that correcting this improves the relationship between genome size and soil pH.

CheckM2 improves MAG quality assessment using machine learning, outperforming its predecessor and handling novel lineages and reduced genomes.