Our Research

We are a team of scientists with diverse academic and cultural backgrounds, working at the intersection between 3D genome biology and cancer genomics. We have a passion for developing innovative methods to advance our understanding of the interplay between genome structure and function, in health and disease.

Our research is driven by the following fundamental questions:

  1. How is chromatin—and chromatin-bound RNAs and proteins—spatially organized in the nucleus of human cells and, collectively, across different cell types, in healthy and diseased tissues? Which forces shape its 3D structure?

  2. How does the spatial organization of chromatin and its temporal dynamics orchestrate fundamental cellular processes, such as DNA replication, repair, and transcription? Vice versa, how do these processes shape chromatin?

  3. How do genomic alterations frequently detected in human cancers, such as somatic copy number alterations (CNAs) and structural variants (SVs), form and evolve over time?

  4. How does the spatial organization of chromatin influence the formation of CNAs and SVs, and, in turn, how is it affected by these alterations?

3D genome biology

Our work in this field aims at characterizing the higher-order organization of chromatin, i.e., where individual parts of the genome are localized in the 3D volume of the nucleus with respect to nuclear landmarks and each other. While 3D genome structure has been thoroughly characterized at the level of chromatin loops, topologically associating domains (TADs), and chromatin compartments, less is known about the higher-order chromatin structure, particularly in cells within their natural tissue context and in tumor cells.

We are especially interested in the radial arrangement of DNA (and nuclear RNAs and proteins) from the nuclear periphery inwards—so-called chromatin radiality. We previously pioneered the GPSeq method for probing the radial position of DNA loci genome-wide at high resolution, and demonstrated that many features of the linear genome (e.g., GC-content) and the epigenome (e.g., histone marks) are distributed along unique radial profiles.

We are now applying GPSeq to compare radial genome organization across different cell and tissue types, and to investigate whether and how chromatin radiality dynamically changes, for example during cellular differentiation or exposure to exogenous stimuli. One exciting application of GPSeq that we are currently pursuing is probing the structure of mitotic chromosomes, which remains rather elusive.

We also developing novel experimental approaches to profile 3D genome structure at the single-cell level as accurately as possible, by combining microscopy with sequencing-based approaches and by developing a single-cell version of GPSeq. Simultaneously, we design novel GPSeq-based assays to probe the radial organization of chromatin-associated RNAs and proteins, driven by the hypothesis that functionally relevant subnuclear RNA/protein gradients might exist.

In parallel to GPSeq, we have developed a comprehensive toolkit, iFISH, for visualization of the spatial organization of DNA and RNAs in the nucleus based on fluorescence in situ hybridization (FISH), using oligonucleotide-based probes. We now leverage iFISH to chart the spatial distribution of nuclear RNAs in different cell types at various differentiation stages, to identify spatial patterns and/or RNA hubs. Furthermore, we apply iFISH to visualize chromosomal territories in normal and tumor cells within their natural tissue context—a largely unexplored area of 3D genome biology.

Cancer genomics

Our work in this field aims at understanding the origin, evolution, and clinical implications of genomic rearrangements in different tumor types and therapeutic contexts. Although thousands of tumors have been profiled by large consortia such as TCGA and PCAWG, we believe that only the ‘tip of the iceberg’ of genomic variation has been surveyed—mainly mutations in coding regions and rearrangements of the linear genome sequence, leaving a vast uncharted territory for further exploration.

For example, very little is known about the spatial arrangement of chromatin in cancer cells, across different regions of the same tumor, and if/how spatial chromatin patterns influence tumor evolution and response to treatments. Similarly, despite tumors being generally considered ‘genomically unstable’, little is known about ongoing genome fragility and DNA damage at the time of diagnosis and during the course of treatment in cancer patients.

We are especially interested in i) understanding how DNA double-strand breaks (DSBs) lead to the formation of CNAs, SVs, and extrachromosomal circular DNA (eccDNA), ii) how these genomic alterations evolve under the selective pressure of therapies, and iii) how they re-shape the higher-order organization of chromatin.

We have previously pioneered the first method—BLESS—for profiling DSBs genome-wide, and over the years we collaborated with many research groups to leverage BLESS and its evolutions (BLISS, sBLISS) for charting the ‘breakome’ in various biological settings. Recently, we have also built our own toolkit for probing CNAs and eccDNAs in tumors, including CUTseq and its single-cell version, scCUTseq, for profiling CNAs, and scCircle-seq for mapping eccDNAs in single cells.

We collaborate with oncologists and pathologists to apply our methods to cohorts of clinically annotated tumor samples, aiming at identifying biomarkers and potentially novel therapeutic targets. Our ultimate goal is to develop clinically interpretable and impactful metrics of the 3D genome and breakome that can be implemented in routine cancer diagnostics. 

To read more about our previous work and tools developed in the lab,
including our new deconvolution software Deconwolf, have a look at Papers.