Panayiotis V. Benos

The paper introduces MORALE, a domain adaptation framework for cross-species TF binding prediction, achieving state-of-the-art performance without adversarial training.

The paper analyzes transcriptome diversity in alcohol-related liver disease to identify novel pathways and genes not detected by differential expression analysis.

HALO models temporal causal relations in single-cell multi-omics data by distinguishing coupled and decoupled latent representations, revealing lineage specification dynamics in AT2 cells.

The paper uses single-cell RNA sequencing to identify pre-vaccination immune cell types, gene expression patterns, and pathways that predict immune response to a preventative MUC1 vaccine.

Single-nucleus multiomic and ex vivo lung slice profiling of human idiopathic pulmonary fibrosis identifies Alveolar Intermediate cells bridging alveolar and Aberrant Basaloid states and reveals AP-1 chromatin injury memory.

The paper identifies transcription factor networks driving fibrotic lung disease using multiomic single-cell data from SSc-ILD and control lungs.

LaGrACE estimates gene program dysregulation using omics data and clinical information to identify molecular subtypes and drug-response signals in diseases.

The paper identifies LEF1 isoforms as key regulators of cellular senescence and aging through transcriptomic analysis and validation in human and mouse models.

The paper identifies a 50-gene high-risk profile predictive of mortality in COVID-19 and IPF, linked to monocyte and T-cell subset imbalances that reverse in post-COVID-19-ILD survivors.

The paper introduces tiSFM, an interpretable neural network for sequence-to-function learning, which outperforms existing models in predicting open chromatin and transcription factor activities.

The paper identifies early fibroblast transitions to a profibrotic state in IPF using scRNA-seq and validates MOXD1 as an early marker.

This study identifies foxm1 as essential for cardiomyocyte proliferation in adult zebrafish after injury, revealing its role in regulating cell cycle genes and enabling regeneration.

Essential Regression infers causal latent factors from multi-omic human datasets, outperforming existing methods in prediction performance and enabling causal inferences.

This paper introduces causalDeepVASE, a deep-learning method for causal inference that identifies both direct and indirect causal relationships in biological systems, outperforming existing methods in simulated and biological data.

ssNPA identifies sample-specific gene network deregulation by learning causal graphs from control data, outperforming existing methods in liver development single-cell RNA-seq analysis.

The paper identifies a gene expression signature for statin sensitivity using transcriptome data from cancer cell lines and validates it experimentally, extending it to TCGA tumors and predicting synergistic drug combinations.