Biology subjects

Eling, N.

Publications and source records attributed to Eling, N..

11 recordsLinked to original sources

bioRxiv2025

Cellohood: multi-granular discovery of cellular neighborhoods with a permutation-invariant set transformer auto-encoder

Discovering cellular neighborhoods and their roles in disease requires computational methods that consider the full breadth of data and offer multi-level granularity and interpretability. Here, we introduce Cellohood, a permutation-invariant set transformer auto-encoder equipped with a clinical association pipeline. Cellohood encodes full readouts for bags of spatially co-localized cells and supports multi-level analyses, providing interpretability by mapping latent dimensions to spatial tissue features. Our model surpasses current methods in accuracy of cellular neighborhood detection for spatial transcriptomics of the human cortex and CODEX spleen data measuring lupus progression. Applied to cancer data across three granularity levels, our method recovers neighborhoods along the expected immune-cold to immune-hot spectrum and further refines them into biologically and clinically meaningful subclasses, revealing spatial patterns linked to prognosis, histology, stage, and tumor mutational burden, and uncovering subgroups that transcend standard classifications. Overall, Cellohood enables in-depth analysis of complex tissues, revealing clinically informative spatial neighborhoods.

bioinformatics↗

bioRxiv2025

Multi-modal image analysis for large scale cancer tissue studies within IMMUcan

MotivationMultiplexed imaging is increasingly used to study tissue architecture in health and disease. To investigate the cancer tumor microenvironment, typically either tissue micro-arrays or small patient cohorts are used to collect and process data. However, studies performed over the course of years, collecting data from thousands of samples are rare and require specialized workflows to ensure sample throughput and reproducibility for data production and processing. Here, we present two such workflows for multiplexed immunofluorescence and imaging mass cytometry of cancer tissues which are applied to a total of roughly 10000 samples from 2500 patients over six years. SummaryIn cancer research, multiplexed imaging allows detailed characterization of the tumor microenvironment (TME) and its link to patient prognosis. The IMMUcan consortium collects multi-modal imaging data from thousands of cancer patients to perform broad molecular and cellular spatial profiling across five cancer indications. Here, we describe two workflows for multiplexed immunofluorescence (mIF) and imaging mass cytometry (IMC) developed within IMMUcan to enable analysis of thousands of cancer tissues. IFQuant supports web-based, user-friendly, and reproducible analysis of mIF data. High sample throughput for IMC is achieved by optimizing experimental protocols and developing a robotic arm for automated slide loading. We provide a resource of 350000 manually labelled cells across 180 cancer samples to accurately annotate cell phenotypes in IMC. All major cell phenotypes and tissue structures correlate well between mIF and IMC. These pipelines form the basis for multiplexed image analysis within IMMUcan and provide computational tools for the larger community.

cancer biology↗

bioRxiv2023

Compressed sensing expands the multiplexity of imaging mass cytometry

The multiplexity of current antibody-based imaging is limited by the number of reporters that can be detected simultaneously. Compressed sensing can be used to recover high-dimensional information from low-dimensional measurements when the data has a structure that allows sparse representation. Previously, in composite in situ imaging (CISI) of transcriptomic data, compressed sensing leveraged the gene co-regulation structure that allows sparse representation and recovered spatial expression of 37 RNA species with the measurement of 11 fluorescent channels. Here, we extended the compressed sensing framework to protein expression data measured by imaging mass cytometry (IMC). CISI-IMC accurately recovered spatial expression of 16 proteins from the images of 8 composite channels, which in effect expanded the current multiplexity limit of IMC by 8 channels. With this ratio, up to 80 protein markers could be compressed into currently available 40 isotope channels. Training the CISI-IMC framework using data collected on tissues from various locations in the human body enabled the decompression of composite data from a wide range of tissue types. Our work laid the foundation for much higher plex protein imaging by using CISI.

bioinformatics↗

bioRxiv2023

cytoviewer: an R/Bioconductor package for interactive visualization and exploration of highly multiplexed imaging data

SummaryHighly multiplexed imaging enables single-cell-resolved detection of numerous biological molecules in their spatial tissue context. Interactive data visualization of multiplexed imaging data is necessary for quality control and hypothesis examination. Here, we describe cytoviewer, an R/Bioconductor package for interactive visualization and exploration of multi-channel images and segmentation masks. The cytoviewer package supports flexible generation of image composites, allows side-by-side visualization of single channels, and facilitates the spatial visualization of single-cell data in the form of segmentation masks. The package operates on SingleCellExperiment, SpatialExperiment and CytoImageList objects and therefore integrates with the Bioconductor framework for single-cell and image analysis. Users of cytoviewer need little coding expertise, and the graphical user interface allows user-friendly navigation. We showcase the functionality of cytoviewer by analysis of an imaging mass cytometry dataset of cancer patients. AvailabilityThe cytoviewer package can be installed from Bioconductor via https://www.bioconductor.org/packages/release/bioc/html/cytoviewer.html. The development version and further instructions can be found on GitHub at https://github.com/BodenmillerGroup/cytoviewer. We provide an R script to exemplify the usage of cytoviewer in the supplementary information. Supplementary informationsSupplementary data are available online.

bioinformatics↗

bioRxiv2021

An end-to-end workflow for multiplexed image processing and analysis

Simultaneous profiling of the spatial distributions of multiple biological molecules at single-cell resolution has recently been enabled by the development of highly multiplexed imaging technologies. Extracting and analyzing biologically relevant information contained in complex imaging data requires the use of a diverse set of computational tools and algorithms. Here, we report the development of a user-friendly, customizable, and interoperable workflow for processing and analyzing data generated by highly multiplexed imaging technologies. The steinbock framework supports image pre-processing, segmentation, feature extraction, and standardized data export. Each step is performed in a reproducible fashion. The imcRtools R/Bioconductor package forms the bridge between image processing and single-cell analysis by directly importing data generated by steinbock. The package further supports spatial data analysis and integrates with tools developed within the Bio-conductor project. Together, the tools described in this workflow facilitate analyses of multiplexed imaging raw data at the single-cell and spatial level.

bioinformatics↗

bioRxiv2021

Multiplexed Imaging Mass Cytometry of Chemokine Milieus in Metastatic Melanoma Characterizes Features of Response to Immunotherapy

Intratumoral immune cells are crucial for tumor control and anti-tumor responses during immunotherapy. Immune cell trafficking into tumors is mediated by chemokines, which are expressed and secreted upon various stimuli and interact with specific receptors. To broadly characterize chemokine expression and function in tumors, we have used multiplex mass cytometry-based imaging of protein markers and RNA transcripts to analyze the chemokine landscape and immune infiltration in metastatic melanoma samples. Tumors that lacked immune infiltration were devoid of most chemokines and exhibited particularly low levels of antigen presentation and inflammation. Infiltrated tumors were characterized by expression of multiple chemokines. CXCL9 and CXCL10 were often localized in patches associated with dysfunctional T cells expressing CXCL13 which was strongly associated with B cell patches and follicles. TCF7+ naive-like T cells, which predict response to immunotherapy, were enriched in the vicinity of B cell patches and follicles. Our data highlight the strength of RNA and protein co-detection which was critical to deconvolve specialized immune microenvironments in inflamed tumors based on chemokine expression. Our findings further suggest that the formation of tertiary lymphoid structures is accompanied by naive and naive- like T cell recruitment, which ultimately boosts anti-tumor activity. One sentence summaryInflammatory chemokine milieus in metastatic melanoma are hotspots of T cell dysfunction and CXCL13 expression, which likely guide the recruitment of B cells and the formation of B cell follicles responsible for anti-tumor immunity.

cancer biology↗

bioRxiv2020

cytomapper: an R/Bioconductor package for visualisation of highly multiplexed imaging data

Highly multiplexed imaging technologies enable spatial profiling of dozens of biomarkers in situ. Standard data processing pipelines quantify cell-specific features and generate object segmentation masks as well as multi-channel images. Therefore, multiplexed imaging data can be visualised across two layers of information: pixel-intensities represent the spatial expression of biomarkers across an image while segmented objects visualise cellular morphology, interactions and cell phenotypes in their microenvironment. Here we describe cytomapper, a computational tool that enables visualisation of pixel- and cell-level information obtained by multiplexed imaging. The package is written in the statistical programming language R, integrates with the image and single-cell analysis infrastructure of the Bioconductor project, and allows visualisation of single to hundreds of images in parallel. Using cytomapper, expression of multiple markers is displayed as composite images, segmentation masks are coloured based on cellular features, and selected cells can be outlined in images based on their cell type, among other functions. We illustrate the utility of cytomapper by analysing 100 images obtained by imaging mass cytometry from a cohort of type 1 diabetes patients and healthy individuals. In addition, cytomapper includes a Shiny application that allows hierarchical gating of cells based on marker expression and visualisation of selected cells in corresponding images. Together, cytomapper offers tools for diverse image and single-cell visualisation approaches and supports robust cell phenotyping via gating.

bioinformatics↗

bioRxiv2018

Altered composition of the γδ T cell pool in lymph nodes during ageing enhances tumour growth

How age-associated decline of immune function leads to increased cancer incidence is poorly understood. Here, we have characterized the cellular composition of the{gamma}{delta} T cell pool in peripheral lymph nodes (pLNs) upon ageing. We found that ageing has minimal cell-intrinsic effects on function and global gene expression of{gamma}{delta} T cells, and TCR{gamma}{delta} diversity remained stable. However, ageing altered TCR{delta} chain usage and clonal structure of{gamma}{delta} T cell subsets. Importantly, IL-17-producing{gamma}{delta} 17 T cells dominated the{gamma}{delta} T cell pool of aged mice - mainly due to the selective expansion of V{gamma}6+{gamma}{delta} 17 T cells and augmented{gamma}{delta} 17-polarisation of V{gamma}4+ T cells. Expansion of the{gamma}{delta} 17 T cell compartment was supported by increased Interleukin-7 expression in the T cell zone of old mice. In a Lewis lung cancer model, pro-tumourigenic V{gamma}6+{gamma}{delta}17 T cells were exclusively activated in the tumour-draining LN and their infiltration into the tumour correlated with increased tumour size in aged mice. Thus, upon ageing, substantial compositional changes of{gamma}{delta} T cell pool in a dysregulated pLN microenvironment promote tumour growth.

immunology↗

bioRxiv2018

Staged developmental mapping and X chromosome transcriptional dynamics during mouse spermatogenesis

Understanding male fertility requires an in-depth characterisation of spermatogenesis, the developmental process by which male gametes are generated. Spermatogenesis occurs continuously throughout a males reproductive window and involves a complex sequence of developmental steps, both of which make this process difficult to decipher at the molecular level. To overcome this, we transcriptionally profiled single cells from multiple distinct stages during the first wave of spermatogenesis, where the most mature germ cell type is known. This naturally enriches for spermatogonia and somatic cell types present at very low frequencies in adult testes. Our atlas, available as a shiny app (https://marionilab.cruk.cam.ac.uk/SpermatoShiny), allowed us to reconstruct the three main processes of spermatogenesis: spermatogonial differentiation, meiosis, and spermiogenesis. Additionally, we profiled the chromatin changes associated with meiotic silencing of the X chromosome, revealing a set of genes specifically and strongly repressed by H3K9me3 in the spermatocyte stage, but which escape post-meiotic silencing in spermatids.

developmental biology↗

bioRxiv2017

Robust expression variability testing reveals heterogeneous T cell responses

Cell-to-cell transcriptional variability in otherwise homogeneous cell populations plays a crucial role in tissue function and development. Single-cell RNA sequencing can characterise this variability in a transcriptome-wide manner. However, technical variation and the confounding between variability and mean expression estimates hinders meaningful comparison of expression variability between cell populations. To address this problem, we introduce a novel analysis approach that extends the BASiCS statistical framework to derive a residual measure of variability that is not confounded by mean expression. Moreover, we introduce a new and robust procedure for quantifying technical noise in experiments where technical spike-in molecules are not available. We illustrate how our method provides biological insight into the dynamics of cell-to-cell expression variability, highlighting a synchronisation of the translational machinery in immune cells upon activation. Additionally, our approach identifies new patterns of variability across CD4+ T cell differentiation.

bioinformatics↗

bioRxiv2017

Whole-body single-cell sequencing of the Platynereis larva reveals a subdivision into apical versus non-apical tissues

Animal bodies comprise a diverse array of tissues and cells. To characterise cellular identities across an entire body, we have compared the transcriptomes of single cells randomly picked from dissociated whole larvae of the marine annelid Platynereis dumerilii1-4. We identify five transcriptionally distinct groups of differentiated cells that are spatially coherent, as revealed by spatial mapping5. Besides somatic musculature, ciliary bands and midgut, we find a group of cells located at the apical tip of the animal, comprising sensory-peptidergic neurons, and another group composed of non-apical neural and epidermal cells covering the rest of the body. These data establish a basic subdivision of the larval body surface into molecularly defined apical versus non-apical tissues, and support the evolutionary conservation of the apical nervous system as a distinct part of the bilaterian brain6.

evolutionary biology↗