Michael I. Jordan

The paper introduces CSDE, a method that improves differential expression analysis in spatial transcriptomics by incorporating expert feedback to mitigate preprocessing biases.

MrVI is a deep generative model that identifies sample-level stratifications and cellular differences without prior cell-type grouping, applied to COVID-19 and IBD data.

This paper introduces a method for functional protein mining with statistical guarantees, improving reliability in homology detection and functional annotation.

This paper introduces VI-VS, a framework for identifying conditionally dependent features in multiomic data while controlling false discovery rates.

MultiVI integrates multi-modal single-cell data by creating a joint representation and imputing missing values while correcting batch effects.

The paper introduces lvm-DE, a Bayesian method for differential expression analysis in scRNA-seq using deep generative models, improving log fold change estimation and differential gene detection.

The paper introduces DestVI, a method for multi-resolution spatial transcriptomics analysis that captures continuous cell state variation, and applies it to study inflammation in lymph nodes and tumors.

scvi-tools provides a Python package for probabilistic analysis of single-cell omics data, offering standardized interfaces and extensible tools for various analysis tasks.

The paper introduces Keris, a framework that uses population-level bulk data to infer single-cell gene expression changes, enabling hypothesis generation without extensive single-cell data collection.

This paper introduces a Bayesian nonparametric framework for identifying individual molecules in super-resolution microscopy time series, using ground-truth data to assess performance.

This paper introduces scVI and scANVI, deep generative models for integrating single-cell RNA-seq data and annotating cell states, showing improved accuracy and scalability over existing methods.

AutoZI detects gene-specific zero-inflation in single-cell RNA sequencing data using a spike-and-slab prior and variational inference, outperforming existing methods on simulated and real data.

scVI maps cells into a low-dimensional latent space with variational inference and neural networks, scaling past a million cells while absorbing batch effects and sensitivity differences.