Arolyn Conwill

AccuSNV is a deep learning tool for high-accuracy bacterial SNV calling that outperforms existing methods in simulated and real datasets.

The paper reveals hidden strain dynamics in skin microbiome species, showing that while strains are transient, some are more stable and may be targets for therapeutic interventions.

The paper introduces a method using Sanger sequencing to rapidly and cheaply measure synthetic bacterial community composition, achieving accurate results in under a day for ~$5 per sample.