Adam Gayoso

AlphaGenome Atlas predicts regulatory effects of non-coding variants across the human genome, improving their prioritization and interpretation.

AlphaGenome predicts functional genomic measurements from DNA sequences with high resolution and accuracy across multiple modalities, outperforming existing methods in variant effect prediction.

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.

Scvi-hub provides a platform for sharing and analyzing single-cell omics datasets using pre-trained models, enabling efficient tasks like visualization and annotation with reduced computational demands.

This paper introduces veloVI, a deep generative model for RNA velocity analysis, showing improved data fit, consistency, and uncertainty quantification compared to existing methods.

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.

PeakVI is a deep generative model for scATAC-seq data analysis that corrects batch effects, preserves biological heterogeneity, and identifies differential accessibility.

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

cell2location uses a Bayesian model to integrate single-cell and spatial transcriptomics, mapping cell types in situ across mouse brain regions and human lymph node immune compartments.

The paper introduces totalVI, a probabilistic framework for joint analysis of RNA and protein data from single cells, demonstrating its utility in integration, dimensionality reduction, clustering, and differential expression testing.

The paper introduces an interpretable variational autoencoder for single-cell RNA-seq to identify gene programs in large datasets.

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.

The paper introduces totalVI, a probabilistic framework for analyzing CITE-seq data by jointly modeling RNA and protein measurements, addressing technical noise and enabling batch correction.

Palantir models cell fate as a probabilistic process using entropy to analyze differentiation trajectories in human hematopoiesis, identifying key transcription factors and terminal states.