Roel Vandepoel

HyDrop v2: Scalable atlas construction for training sequence-to-function models

The paper introduces HyDrop v2, a microfluidic system for generating large scATAC-seq atlases, and shows its data is comparable to commercial platforms for training sequence-to-function models.

System-wide extraction of cis-regulatory rules from sequence-to-function models in human neural development

The paper extracts cis-regulatory rules from sequence-to-function models in human neural development using new single-cell multiome data and a computational tool called TF-MINDI.

Enhancer-driven cell type comparison reveals similarities between the mammalian and bird pallium

The paper compares enhancer codes of cell types between mammalian neocortex and avian pallium using deep learning, finding higher divergence in excitatory neurons than other cell types.

Nova-ST: Nano-Patterned Ultra-Dense platform for spatial transcriptomics

Nova-ST is a cost-effective, high-resolution spatial transcriptomics platform that outperforms existing methods in sensitivity for mouse brain sections.

Cell type directed design of synthetic enhancers

This study uses deep learning to design synthetic enhancers that target specific cell types in fruit flies and humans, revealing the enhancer code as optimal TF activator motifs and absence of repressor motifs.

Enhancer grammar of liver cell types and hepatocyte zonation states

The paper maps enhancer gene regulatory networks (GRNs) in liver cell types and identifies TFs governing hepatocyte zonation states using multiomics and deep learning.