Min Oh

MetaCompare 2.0 differentiates human health and ecological resistome risks using metagenomic data, showing higher risk scores for human-contaminated environments.

This paper introduces DeepGeni, an interpretable autoencoder for microbiome analysis, showing improved prediction of melanoma immunotherapy response and identifying key microbial taxa.

This study introduces DeepBioGen, a deep generative model for augmenting sequencing profiles to improve prediction model generalizability on unseen data.

DeepMicro uses deep representation learning to predict diseases from microbiome data, outperforming existing methods and achieving faster training.