Tommi Vatanen

LGTM couples a neural topic model with additive basis-function Gaussian processes, learning microbial co-abundance topics and how covariates relate to their trajectories in longitudinal gut microbiome data.

The paper identifies gut bacteria-produced metabolites that inhibit HSD11B2, altering cortisol signaling and linking microbiome activity to blood pressure regulation.

The paper identifies new phosphorothioate DNA modification systems in the human gut microbiome using genomic and mass spectrometric analysis.

CAZyLingua uses protein language models to annotate carbohydrate-active enzymes in metagenomic data, outperforming traditional methods and revealing new annotations in disease-related datasets.

The paper investigates how industrialization affects microbial and physiological traits in the human gut across diverse populations, finding homogenized microbiomes, reduced diversity, and impaired disease prediction accuracy in less industrialized groups.

The paper develops a metagenome analysis workflow combining de novo genome reconstruction, taxonomic profiling, and deep learning-based functional annotation, showing improved coverage over existing methods.

The paper predicts structures and annotates functions for diverse microbial protein sequences, identifying novel folds and highlighting the saturation of structural space.

This paper introduces a Graph Convolutional Network for predicting protein function and identifying important residues, showing improved accuracy over sequence-based methods.

LonGP is a Gaussian process regression model for longitudinal studies that handles time-varying covariates and non-linear effects, showing performance on simulated and real -omics data.