Daria Doncevic

The paper benchmarks perturbation prediction tools in single-cell RNA-seq using scArchon, revealing performance disparities and highlighting the need for gene-level evaluation.

The paper introduces COBRA, an interpretable deep learning model for scRNA-seq data that disentangles covariate effects, and demonstrates its application on interferon-stimulated mouse datasets.

This paper introduces OntoVAE, a VAE that incorporates ontologies to predict genetic and drug-induced perturbations using bulk and single-cell transcriptomic data.

DECONbench benchmarks deconvolution methods for quantifying tumor cell-type heterogeneity by providing datasets, tools, and performance evaluation.