Gerald Quon

The paper integrates multiple datasets to create a comprehensive human liver cell atlas, identifying new cell types and associations with demographic factors.

CELLECTION predicts emergent phenotypes from single-cell populations by associating subgroups with different phenotypes across diverse tasks.

The paper identifies human-specific gene expansions linked to brain evolution by analyzing genome sequences and functional models in zebrafish.

This paper introduces a conditional latent diffusion model to separate batch effects from morphological features in zebrafish imaging data, enabling accurate genotype classification.

The paper develops a framework to infer ERK signaling dynamics from fixed-cell immunostains by combining machine learning and differential equations, showing that protein levels reflect different activation timescales.

sciLaMA integrates gene embeddings from large language models with scRNA-seq data using a paired-VAE architecture, improving downstream tasks like batch correction and clustering while maintaining computational efficiency.

The paper identifies sex-specific transcriptomic changes in a mouse model of Rett syndrome, showing more differentially expressed genes in females and non-cell-autonomous effects.

The paper identifies subclass-specific genes associated with electrophysiological features in mouse interneurons and links cell adhesion molecules to schizophrenia risk.

scProjection maps multi-modal RNA measurements to scRNA-seq atlases to extract single cell gene expression profiles and improve spatial and multi-modal analysis.

This paper introduces an interpretable variational autoencoder for genomics that identifies gene neighborhoods and hubs, revealing correlations between neuronal differentiation and mitochondrial complex co-expression.

scBFA mitigates technical noise in single cell genomics data by analyzing detection patterns instead of quantification measurements, improving cell type identification and trajectory inference.

The paper compares cell types in human and mouse cortex using single nucleus RNA-sequencing, finding conserved architecture with species-specific differences.

scAlign integrates scRNA-seq datasets and identifies rare cell populations, demonstrating its ability to detect a rare cell type associated with malaria transmission.

CONVERGE identifies co-localization of GWAS variants with transcription factor binding sites in regulatory regions to find novel trait regulators, outperforming existing methods in validating type 2 diabetes regulators.

The paper shows that EGFR/Ras/ERK signaling in invasive breast cancer cells produces more dynamic and disordered ERK activity, leading to heterogeneous gene expression compared to non-malignant cells.