Daniel Paysan


Last Name

Paysan

First Name

Daniel

ORCID

0000-0003-4995-7527

Organisational unit

Filters

2023 - 20256
Reset filters

Search Results

Publications 1 - 6 of 6
  • Image2Reg: Linking chromatin images to gene regulation using genetic and chemical perturbation screens
    Item type: Journal Article
    Paysan, Daniel; Radhakrishnan, Adityanarayanan; Zhang, Xinyi; et al. (2025)
    Cell Systems
    Representation learning provides an opportunity to uncover the link between 3D genome organization and gene regulatory networks, thereby connecting the physical and the biochemical space of a cell. Our method, Image2Reg, uses chromatin images obtained in large-scale genetic and chemical perturbation screens. Through convolutional neural networks, Image2Reg generates gene embedding that represents the effect of gene perturbation on chromatin organization. In addition, combining protein-protein interaction data with cell-type-specific transcriptomic data through a graph convolutional network, we obtain a gene embedding that represents the regulatory effect of genes. Finally, Image2Reg learns a map between the resulting physical and biochemical representation of cells, allowing us to predict the perturbed gene modules based on chromatin images. Our results confirm the deep link between chromatin organization and gene regulation and demonstrate that it can be harnessed to identify drug targets and genes upstream of perturbed phenotypes from a simple and inexpensive chromatin staining.
  • Detecting radio- and chemoresistant cells in 3D cancer co-cultures using chromatin biomarkers
    Item type: Journal Article
    Pekeč, Tina; Venkatachalapathy, Saradha; Shim, Anne R.; et al. (2023)
    Scientific Reports
    The heterogenous treatment response of tumor cells limits the effectiveness of cancer therapy. While this heterogeneity has been linked to cell-to-cell variability within the complex tumor microenvironment, a quantitative biomarker that identifies and characterizes treatment-resistant cell populations is still missing. Herein, we use chromatin organization as a cost-efficient readout of the cells’ states to identify subpopulations that exhibit distinct responses to radiotherapy. To this end, we developed a 3D co-culture model of cancer spheroids and patient-derived fibroblasts treated with radiotherapy. Using the model we identified treatment-resistant cells that bypassed DNA damage checkpoints and exhibited an aggressive growth phenotype. Importantly, these cells featured more condensed chromatin which primed them for treatment evasion, as inhibiting chromatin condensation and DNA damage repair mechanisms improved the efficacy of not only radio- but also chemotherapy. Collectively, our work shows the potential of using chromatin organization to cost-effectively study the heterogeneous treatment susceptibility of cells and guide therapeutic design.
  • Computational Approaches Linking Chromatin Organization and Genome Regulation
    Item type: Doctoral Thesis
    Paysan, Daniel (2024)
    In cells, the often meter-long DNA is packed into nuclei that are only a few micrometer in size. The packing of the DNA results in a highly structured organization of the genome, i.e. the 3D chromatin organization. This organization varies between cell states and is closely related to the gene regulatory networks that control cellular function in cells. The connection arises from the optimized packing, which for instance is expressed by the spatial colocalization of genes for their efficient co-regulation. However, the depth of the connection between the chromatin organization and genome regulation, along with its regulating mechanisms, is not yet fully understood. It also remains unclear whether this connection is profound enough to use the chromatin organization of cells as an unified readout of functional cell states and study cell state transitions. This would have wide-ranging implications for disease diagnostics and therapeutic design. In this thesis, we address these important questions. Specifically, we first demonstrate that deep representation learning can be used to effectively model the link between chromatin organization and gene regulation. To achieve this, we developed a computational pipeline that learns how the gene regulatory programs of cells are reflected in their chromatin organization. This enables our pipeline to accurately identify alterations in these regulatory programs simply by using cost-efficient chromatin images as a readout of their chromatin organization. Our work provides novel avenues for studying gene regulation and alterations thereof using image-based readouts of the nuclear chromatin organization and thus emphasizes the deep connection between chromatin organization and genome regulation. While it has become evident that the functional optimization of the 3D chromatin packing gives rise to this connection, the mechanisms responsible for the optimized packing are not well understood. This particularly applies to the establishment of contacts between genomic regions from different chromosomes, which are crucial for cell function. In this thesis, we propose that the binding of long noncoding RNAs (lncRNAs) to the DNA may be a mechanism facilitating the formation and stability of these contacts, thereby enabling the colocalization and co-regulation of important gene clusters. To this end, we analyze multimodal sequencing data to identify spatial gene clusters formed by interchromosomal contacts. We show that the formation or loss of these gene clusters during aging could be explained by the changing abundance of specific lncRNAs. These lncRNAs might act as “glues” binding the corresponding genomic regions together. Although additional validation experiments are ongoing, our findings provide convincing evidence that the transcriptional control of lncRNAs and their binding to the DNA plays an important role for the functionally optimized 3D genome organization. Finally, we demonstrate how computational methods can be used to exploit the link between chromatin organization and genome regulation, enabling the development of cost-effective image-based chromatin biomarkers with applications in disease diagnostics and therapeutic design. We first showcase this using chromatin images of immune cells obtained from liquid biopsies of tumor patients. These images are input to a computational pipeline which we developed and which uses simple image analysis and machine learning methods. We show that this pipeline identifies chromatin biomarkers of immune cells that enable accurate tumor diagnosis. We further illustrate the potential of such chromatin biomarkers by applying a similar pipeline to characterize differences in the chromatin organization of distinct B cell populations in lymphoid tissue. Our corresponding analyses reveal chromatin condensation differences in these B cell populations, which correlate with the presence or absence of other immune cell populations in their local microenvironments. This could potentially provide novel insights into the mechanisms that control T cell infiltration in B cell lymphomas, also implying that inhibition of chromatin condensation may prevent immune cell evasion in B cell lymphomas. Thus, this study also highlights the potential of chromatin biomarkers to guide therapeutic design. In summary, in this thesis we present novel computational methods to study, describe, model, and leverage the relationship between chromatin organization and genome regulation. The results presented in this thesis improve our understanding of this connection and its regulating mechanisms as well as demonstrate its large potential for applications such as disease diagnostics and therapy evaluation.
  • Imaging and AI based chromatin biomarkers for diagnosis and therapy evaluation from liquid biopsies
    Item type: Journal Article
    Challa, Kiran; Paysan, Daniel; Leiser, Dominic; et al. (2023)
    npj Precision Oncology
    Multiple genomic and proteomic studies have suggested that peripheral blood mononuclear cells (PBMCs) respond to tumor secretomes and thus could provide possible avenues for tumor prognosis and treatment evaluation. We hypothesized that the chromatin organization of PBMCs obtained from liquid biopsies, which integrates secretome signals with gene expression programs, provides efficient biomarkers to characterize tumor signals and the efficacy of proton therapy in tumor patients. Here, we show that chromatin imaging of PBMCs combined with machine learning methods provides such robust and predictive chromatin biomarkers. We show that such chromatin biomarkers enable the classification of 10 healthy and 10 pan-tumor patients. Furthermore, we extended our pipeline to assess the tumor types and states of 30 tumor patients undergoing (proton) radiation therapy. We show that our pipeline can thereby accurately distinguish between three tumor groups with up to 89% accuracy and enables the monitoring of the treatment effects. Collectively, we show the potential of chromatin biomarkers for cancer diagnostics and therapy evaluation.
  • Unsupervised representation learning of chromatin images identifies changes in cell state and tissue organization in DCIS
    Item type: Journal Article
    Zhang, Xinyi; Venkatachalapathy, Saradha; Paysan, Daniel; et al. (2024)
    Nature Communications
    Ductal carcinoma in situ (DCIS) is a pre-invasive tumor that can progress to invasive breast cancer, a leading cause of cancer death. We generate a large-scale tissue microarray dataset of chromatin images, from 560 samples from 122 female patients in 3 disease stages and 11 phenotypic categories. Using representation learning on chromatin images alone, without multiplexed staining or high-throughput sequencing, we identify eight morphological cell states and tissue features marking DCIS. All cell states are observed in all disease stages with different proportions, indicating that cell states enriched in invasive cancer exist in small fractions in normal breast tissue. Tissue-level analysis reveals significant changes in the spatial organization of cell states across disease stages, which is predictive of disease stage and phenotypic category. Taken together, we show that chromatin imaging represents a powerful measure of cell state and disease stage of DCIS, providing a simple and effective tumor biomarker.
  • Aggressive B cell lymphomas retain ATR-dependent determinants of T cell exclusion from the germinal center dark zone
    Item type: Journal Article
    Cancila, Valeria; Bertolazzi, Giorgio; Chan, Allison S.Y.; et al. (2025)
    The Journal of Clinical Investigation
    The germinal center (GC) dark zone (DZ) and light zone represent distinct anatomical regions in lymphoid tissue where B cell proliferation, immunoglobulin diversification, and selection are coordinated. Diffuse large B cell lymphomas (DLBCLs) with DZ-like gene expression profiles exhibit poor outcomes, though the reasons are unclear and are not directly related to proliferation. Physiological DZs exhibit an exclusion of T cells, prompting exploration of whether T cell paucity contributes to DZ-like DLBCL. We used spatial transcriptomic approaches to achieve higher resolution of T cell spatial heterogeneity in the GC and to derive potential pathways that underlie T cell exclusion. We showed that T cell exclusion from the DZ was linked to DNA damage response (DDR) and chromatin compaction molecular features characterizing the spatial DZ signature, and that these programs were independent of activation-induced cytidine deaminase (AID) activity. As ATR is a key regulator of DDR, we tested its role in the T cell inhibitory DZ transcriptional imprint. ATR inhibition reversed not only the DZ transcriptional signature, but also DZ T cell exclusion in DZ-like DLBCL in vitro microfluidic models and in in vivo samples of murine lymphoid tissue. These findings highlight that ATR activity underpins a physiological scenario of immune silencing. ATR inhibition may reverse the immune-silent state and enhance T cell–based immunotherapy in aggressive lymphomas with GC DZ–like characteristics.
Publications 1 - 6 of 6