Maude M. David

This paper introduces Nammu, a state-space model for metagenomic data that outperforms Transformer-based models in taxonomic and functional predictions.

This paper introduces BioGeoFormer, a protein language model that classifies unknown genes into biogeochemical pathways, outperforming traditional methods in identifying functional roles in deep-sea microbial genes.

This study uses topic modeling to integrate multi-omic data from the gut microbiome of children with and without autism, identifying microbial processes associated with autism.

This paper introduces a method to analyze microbiome data using word-embedding techniques to capture community-level properties, showing improved predictive power for inflammatory bowel disease classification.