Research

Research Overview

Our main goal is to unravel the diverse mechanisms of mammalian gene regulation and their roles in development and disease. Gene activation is controlled locally through transcription factors, chromatin modifiers, and the transcription and RNA processing machineries. Gene expression is also intricately regulated by the three-dimensional (3D) structure of the genome, which modifies the location of genes relative to activating or repressing biochemical environments. Repressing nuclear locations include the nuclear lamina and heterochromatin. Activating 3D genome conformations can work by regulating the physical proximity between regulatory regions and their responsive genes. As many disease-associated genetic variants fall in non-coding regulatory elements scattered in the genome, our work aims to advance the functional interpretation of the linear genome sequence. We especially put effort in devising experimental and computational approaches to map 3D genome structure in different cell types, in development and in disease, to study developmental and disease mechanisms, towards novel diagnostics, prognostics and therapeutics.

We have developed Genome Architecture Mapping (GAM), an orthogonal 3D genome folding mapping technology which yields fine maps of chromatin contacts from small cell numbers (500−1000), and is uniquely powerful to quantify different metrics of 3D genome structure, such as multiway contacts and chromatin melting. With immuno-GAM, we introduced the selection of cell types within complex tissues to enable the application of GAM in rare cells, such as dopaminergic neurons in the midbrain. GAM is an inherently suited platform for multimodal molecular phenotyping of biological samples, from genome sequence and 3D structure, to transcript and protein quantification. As a first step towards GAM multimodality, we have developed Oligo-Seq, a novel approach to detect minute amounts of specific RNA sequences within thin tissue cryosections, as used for GAM. We currently develop the application of Oligo-seq in spatial transcriptomics, especially in archival clinical samples (e.g. FFPE). By further developing multimodal integration of 3D genome topology, with the abundance of specific regulatory factors and gene expression, at the single-cell level, we aim to extract dynamic states of gene regulation and cause-effect relationships that connect 3D genome structure with gene expression. Ultimately, we seek to advance the interpretability of the linear genome sequence, to gain a deeper understanding of 3D genome regulation mechanisms in gene expression, and to advance their predictability.