David M. Blei

Decipher is a deep generative model that integrates single-cell RNA-seq data to visualize and characterize transitions between normal and perturbed cell states across different diseases.

The paper introduces VI-MIDAS, a probabilistic framework for analyzing microbiome data, and applies it to Tara Ocean data to identify microbial community modules and associations.

Starfysh integrates spatial transcriptomics and histology to analyze tumor microenvironments, identifying disease-specific spatial hubs and immune interactions in breast cancers.

A hierarchical state space model splits optical neural recordings into linear segments and pools partial data across worms, recovering discrete states whose transitions match hand-labelled crawling, reversals and turns.

This paper introduces scHPF, a Bayesian method for discovering gene signatures from scRNA-seq without prior normalization, and finds spatially biased expression patterns in glioma regions linked to poor survival.

The paper introduces the Markov link method to predict measurements from different experimental techniques and evaluates it on single-cell RNA sequencing data.

This paper introduces HTFA, a Bayesian hierarchical matrix factorization model, to discover dynamic full-brain functional connectivity patterns from fMRI data and shows it can decode cognitive states during storytelling and TV watching.