Liesbeth Minnoye

The paper identifies and characterizes enhancer activity in melanoma cell states using MPRA and DeepMEL2, showing AP-1 drives MES enhancers and SOX/MITF influence MEL enhancers.

The paper uses a deep learning model on melanoma genomes to prioritize enhancer mutations by analyzing allelic imbalance in chromatin accessibility and gene expression.

The paper uses deep learning to analyze enhancer logic across species in melanoma, identifying conserved regulatory complexes and their roles.

The paper shows DNA methylation prevents HIF binding, affecting hypoxia responses and tumour immunotolerance.

The paper identifies an intermediate cell state in melanoma through single-cell RNA-seq and ATAC-seq, revealing a gene regulatory network that governs phenotype switching and potential therapeutic targets.

The paper introduces cisTopic, a probabilistic framework for discovering cis-regulatory topics and cell states from single-cell epigenomics data, and validates its findings using RNA-seq co-expression networks and perturbation experiments.