Biology subjects

Campbell, K. R.

Publications and source records attributed to Campbell, K. R..

At least 19 recordsLinked to original sources

bioRxiv2026

An Interpretable 3D Bag-Of-Visual-Words Pipeline for Volumetric Microscopy Classification

Fluorescence microscopy increasingly produces complex volumetric datasets whose biologically meaningful differences are difficult to capture with hand-crafted measurements, especially when signal is distributed across three-dimensional space. Here, we present an interpretable 3D Bag-of-Visual-Words (BoVW) pipeline for classification and analysis of volumetric microscopy data. The framework detects multiscale local keypoints, computes rotationally robust 3D gradient-based descriptors, and aggregates them into image-level visual-word representations. These features are then used for low-dimensional visualization and logistic regression classification, while model weights are mapped back to the original volumes to generate attention maps that localize discriminative structures. We applied the pipeline to two cerebellar granule neuron datasets spanning both ideal and non-ideal imaging conditions. In a near-isotropic lattice light-sheet dataset of chromatin organization, the method separated control and NIPBL loss-of-function nuclei and supported accurate classification, with strongest performance in the facultative heterochromatin and H3.3 channels. Attention mapping and downstream connected-component and Haralick analyses revealed that loss-of-function nuclei contained more fragmented high-attention regions and smoother, more homogeneous chromatin-associated textures than controls. We then evaluated the same framework on an anisotropic confocal timelapse dataset of receptor clustering in dense neuronal cultures, where single-cell segmentation was impractical. Despite these challenges, the representation captured the expected ligand-driven clustering response and resolved subtler differences associated with a polarity protein overexpression. Together, these results establish a simple, interpretable, and broadly applicable framework for extracting biologically meaningful structure from volumetric microscopy datasets while preserving native 3D context.

cell biology↗

bioRxiv2025

Rakaia: interactive discovery of spatial biology at scale

Spatial biology data throughput currently outpaces interpretable analysis, limiting large-scale discovery and translation. Here we present Rakaia, a browser-based platform for multiplexed imaging and spatial transcriptomics, empowering code-free interactive exploration and evaluation. Rakaia enables visualization, annotation, and feature-based querying with prioritization across thousands of images. We use Rakaia to identify cell types of interest in >200 highly multiplexed images from non-malignant human breast tissue. Rakaia is available at: https://rakaia.io/

bioinformatics↗

bioRxiv2025

Integrated spatial proteomics of human PDAC uncovers an expanded tumour-immune-stroma spectrum with genomic associations

Distinctively, pancreatic ductal adenocarcinoma (PDAC) consists of sparse tumour lesions intertwined with extensive desmoplastic stroma. The complexity of tumour-microenvironment interactions within this desmoplasia poses a challenge for accurate tumour profiling and patient stratification, and characterizes a profoundly chemoresistant tumour. Here we mapped the spatial relationships between tumour, stroma, and immune cell compartments delineating tumour and microenvironment types that expand the classical to basal spectrum of human PDAC. We used imaging mass cytometry to profile the in situ multi-cellular organization of 81 cell types in resected cases with paired whole genome sequencing. Cell types, functions, and pathway activation were distributed as highly reproducible environments in discrete locations throughout these tumours, which we deep-profiled using laser-capture mass spectrometry. We show that the connections between tumour phenotypes, vascularization, immune response, and stromal biophysical state are reinforced by genomic aberrations, altered by treatment, and associated with patient outcome. Predictive machine-learning models showed that spatial single cell data outperformed genomic or clinical features but integrated multi-omics models provide the best prediction of patient survival with compressed models requiring only 10 non-redundant robust molecular measures associated with the phenotypic spectrum of PDAC. Together, these findings define a phenotypic and molecular framework of PDAC that captures tumour-microenvironment co-dependencies and offers a refined basis for patient stratification and therapeutic targeting.

cancer biology↗

bioRxiv2025

Learning copy number dependent variation in single tumour cell transcriptomes with deep generative models

Large-scale copy number aberrations are a hallmark of many cancers, showing widespread associations with tumour cell phenotypes, therapy resistance, and patient outcomes. Several existing methods can infer copy number profiles from single-cell RNA-sequencing data. However, there is currently no methodology to separate the effects of the copy number state on single tumour cell phenotypes from the copy number-independent variation. We present ISOMERIC, an unsupervised deep learning framework to learn disentangled representations of single-cell expression data that are dependent on or independent of copy number profiles. We apply ISOMERIC to multiple cancer types and examine how copy number shapes transcriptional states. We find distinct associations between copy number-dependent and independent variation and clinical subtypes in multiple cancer types and link such variation to tumour microenvironment phenotypes. Together, this establishes a principled framework for untangling the effects of genetic and intrinsic variation on tumour transcriptomes and the consequences on clinically meaningful phenotypes.

bioinformatics↗

bioRxiv2025

Multi-modal Disentanglement of Spatial Transcriptomics and Histopathology Imaging

Spatially-resolved expression profiling data has revolutionized biological research with multiple emerging clinical applications. Spatial transcriptomic assays are often jointly measured with histopathology imaging data, which is frequently used for diagnosing and staging various diseases. However, determining the extent to which the spatial transcriptomic and histopathology data represent overlapping or unique sources of variation is challenging, particularly given the myriad of factors influencing both, including expression variation, spatial context, tissue morphology, and batch effects. Here, we view this challenge as multi-modal disentanglement and develop an evaluation framework. We introduce SpatialDIVA, a disentanglement technique for jointly measured spatially resolved transcriptomics and histopathology data. We demonstrate that SpatialDIVA outperforms baseline techniques in disentangling salient factors of variation in curated pathologist-annotated multi-sample colorectal and pancreatic cancer cohorts. Further, SpatialDIVA removes batch effects from multi-modal data, allows for factor covariance analysis, and yields actionable biological insights through a novel conditional multi-modal generation method. The SpatialDIVA model, evaluation code, and datasets are available at https://github.com/hsmaan/SpatialDIVA.

genomics↗

bioRxiv2024

Interactive design and validation of antibody panels using single-cell RNA-seq atlases

Single-cell RNA-sequencing holds promise for identifying novel markers of cellular variation for antibody-based technologies. However, antibody panel design is often difficult due to multiple experimental and biological constraints. We introduce Cytomarker, an interactive platform enabling human-in-the-loop design of antibody panels from single-cell transcriptomic data. We use Cytomarker to spatially profile human mammary tissue subpopulations and develop a novel antibody screening approach to validate granular subpopulation predictions across >3.5M cells.

bioinformatics↗

bioRxiv2024

Computational design and evaluation of optimal bait sets for scalable proximity proteomics

The spatial organization of proteins in eukaryotic cells can be explored by identifying nearby proteins using proximity-dependent biotinylation approaches like BioID. BioID defines the localization of thousands of endogenous proteins in human cells when used on hundreds of bait proteins. However, this high bait number restricts the approachs usage and gives these datasets limited scalability for context-dependent spatial profiling. To make subcellular proteome mapping across different cell types and conditions more practical and cost-effective, we developed a comprehensive benchmarking platform and multiple metrics to assess how well a given bait subset can reproduce an original BioID dataset. We also introduce GENBAIT, which uses a genetic algorithm to optimize bait subset selection, to derive bait subsets predicted to retain the structure and coverage of two large BioID datasets using less than a third of the original baits. This flexible solution is poised to improve the intelligent selection of baits for contextual studies.

bioinformatics↗

bioRxiv2024

Decoding multicellular niche formation in the tumour microenvironment from nonspatial single-cell expression data

Accurate inference of granular cell states that co-occur within the tumour microenvironment (TME) is central to defining pro- and anti-tumour environments. Here, we describe how ecosystems of cell populations can be robustly inferred from nonspatial single-cell RNA-seq (scRNA-seq) atlases. Leveraging a unique discovery-validation setup across eight scRNA-seq datasets profiling pancreatic ductal adenocarcinoma (PDAC), we show highly consistent co-occurrence of fine-grained cell states across patients and characterize the positive predictive value of such analyses. Building on this, we develop a novel probabilistic model to quantify multi-cellular ecosystems directly from such atlas-scale scRNA-seq datasets. By mapping these ecosystems to spatial transcriptomics data, we demonstrate that such ecosystems represent bona fide spatial variation within individual tumours despite being learned from nonspatial data. Importantly, through mapping these ecosystems across two large clinical cohorts, we show they are more predictive of therapy response than any individual cell state and are associated with specific tumour somatic mutations. Together, this work lays the foundation for inferring reproducible multicellular ecosystems directly from large nonspatial scRNA-seq atlases.

bioinformatics↗

bioRxiv2024

Spatial proteomic mapping of human nuclear bodies reveals new functional insights into RNA regulation

Nuclear bodies are diverse membraneless suborganelles with emerging links to development and disease. Explaining their structure, function, regulation, and implications in human health will require understanding their protein composition; however, isolating nuclear bodies for proteomic analysis remains challenging. We present the first comprehensive proximity proteomics-based map of nuclear bodies, featuring 140 bait proteins (encoded by 119 genes) and 1,816 unique prey proteins. We identified 641 potential nuclear body components, including 131 paraspeckle proteins and 147 nuclear speckle proteins. After validating 31 novel paraspeckle and nuclear speckle components, we discovered regulatory functions for the poorly characterised nuclear speckle- and RNA export-associated proteins PAXBP1, PPIL4, and C19ORF47, and revealed that QKI regulates paraspeckle size. This work provides a systematic framework of nuclear body composition in live cells that will accelerate future research into their organisation and roles in human health and disease.

cell biology↗

bioRxiv2024

Segmentation error aware clustering for highly multiplexed imaging

Spatial protein expression technologies can map cellular content and organization by simultaneously quantifying the expression of >40 proteins at subcellular resolution within intact tissue sections and cell lines. However, necessary image segmentation to single cells is challenging and error prone, easily confounding the interpretation of cellular phenotypes and cell clusters. To address these limitations, we present STARLING, a novel probabilistic machine learning model designed to quantify cell populations from spatial protein expression data while accounting for segmentation errors. To evaluate performance we developed a comprehensive benchmarking workflow by generating highly multiplexed imaging data of cell line pellet standards with controlled cell content and marker expression and additionally established a novel score to quantify the biological plausibility of discovered cellular phenotypes on patient derived tissue sections. Moreover, we generate spatial expression data of the human tonsil - a densely packed tissue prone to segmentation errors - and demonstrate cellular states captured by STARLING identify known cell types not visible with other methods and enable quantification of intra- and inter- individual heterogeneity. STARLING is available at https://github.com/camlab-bioml/starling.

bioinformatics↗

bioRxiv2024

Beyond benchmarking: towards predictive models of dataset-specific single-cell RNA-seq pipeline performance

The advent of single-cell RNA-sequencing (scRNA-seq) has driven significant computational methods development for all steps in the scRNA-seq data analysis pipeline, including filtering, normalization, and clustering. The large number of methods and their resulting parameter combinations has created a combinatorial set of possible pipelines to analyze scRNA-seq data, which leads to the obvious question: which is best? Several benchmarking studies have sought to compare methods to answer this, but frequently find variable performance depending on dataset and pipeline characteristics. Alternatively, the large number of publicly available scRNA-seq datasets along with advances in supervised machine learning raise a tantalizing possibility: could the optimal pipeline be predicted for a given dataset? Here we begin to answer this question by applying 288 scRNA-seq analysis pipelines to 86 datasets and quantifying pipeline success via a range of measures evaluating cluster purity and biological plausibility. We build supervised machine learning models to predict pipeline success given a range of dataset and pipeline characteristics. We find both that prediction performance is significantly better than random and that in many cases pipelines predicted to perform well provide clustering outputs similar to expert-annotated cell type labels. Finally, we identify characteristics of scRNA-seq datasets that correlate with strong prediction performance that could guide when such prediction models may be useful.

bioinformatics↗

bioRxiv2023

Single cell decoding of drug induced transcriptomic reprogramming in triple negative breast cancers

BackgroundThe encoding of cell intrinsic resistance states in breast cancer reflects the contributions of genomic and non-genomic variation. However, identifying the potential contributions of each requires accurate measurement and subtraction of the contribution of clonal fitness from co-measurement of transcriptional states. Somatic genomic variation in gene dosage, copy number variation, is the dominant mutational mechanism in breast cancer contributing to transcriptional variation and has recently been shown to contribute to platinum chemotherapy resistance states. Here we deploy time series measurements of triple negative breast cancer single cell transcriptomes in conjunction with co-measured single cell copy number associated clonal fitness to identify the contributions of genomic and non-genomic mechanisms to drug associated transcription states. ResultsWe generated serial scRNA-seq data (126,556 cells) from triple negative breast cancer (TNBC) patient-derived xenograft (PDX) experiments over 2.5 years in duration, and matched it against genomic copy number single cell data from the same biological samples. We show that the cell memory of transcriptional states of TNBC tumors serially exposed to platinum identifies distinct clonal responses within individual tumours. Copy-number clones with high drug fitness leading to clonal sweeps exhibit less transcriptional reversion, whereas clones with weak drug fitness exhibit highly dynamic transcription on drug withdrawal. Pathway analysis shows that copy number associated and copy number independent transcripts converge on epithelial-mesenchymal transition (EMT) and cytokine signaling states associated with resistance. We show from trajectory analysis that transcriptional reversion exhibits hysteresis, indicating that new intermediate transcriptional states are generated by platinum exposure. ConclusionsWe discovered that copy number clones with strong genotype associated fitness under platinum became fixed in their states, resulting in minimal transcriptional reversion on drug withdrawal. In contrast clones with weaker fitness undergo non-genomic transcriptional plasticity and these distinct responses co-exist within single tumours. Together the data suggest that copy number associated and copy number independent transcriptional states may contribute to platinum drug resistance within individual tumours. The dominance of genomic or non-genomic mechanisms within individual polyclonal tumours has implications for approaches to restoration of drug sensitivity and re-treatment strategies. Data availabilityUploaded Data URL: https://ega-archive.org/studies/EGAS00001007242 Github manuscript: https://github.com/molonc/drug_resistant_material/

genomics↗

bioRxiv2023

An Engineered 3D Co-culture Model of Primary Macrophages and Patient-Derived Tumour Cells to Explore Cellular Responses in the Graded Hypoxic Microenvironment of Pancreatic Cancer

In pancreatic ductal adenocarcinoma (PDAC), tumour associated macrophages (TAMs) are a heterogeneous immune cell population that interact with cancer cells to promote malignancy, chemo-resistance, and immunosuppression. Aside from TAMs, hypoxia is a prominent feature of PDAC that can rewire cells to survive and enhance malignancy in the tumour microenvironment (TME). Deciphering the interactions between macrophages, cancer cells and hypoxia could lead to the development of effective immune-targeted therapies for PDAC. However, there are only a few models that physiologically recapitulate the PDAC TME and allow for meaningful interrogation of cancer-immune cell interactions in hypoxia. Here, we develop a model of primary macrophages and PDAC patient organoid-derived cells by adapting TRACER, a paper-based, engineered 3D model that allows snapshot analysis of cellular response in hypoxia. In this study, we establish a direct co-culture method of primary macrophages and PDAC organoid cells in TRACER and demonstrate that TRACER co-cultures generate hypoxic gradients and show expected phenotypic responses to this hypoxic gradient. Moreover, we report for the first time in a human in vitro model that hypoxic macrophages exert a graded chemoprotective effect on gemcitabine-treated PDAC organoid cells, and that interactions between cancer cells and macrophages from the inner layers of TRACER indirectly attenuate the inflammatory response of donor-derived T-cells. Overall, the TRACER co-culture system is a novel, fully human 3D in vitro cancer-immune model for evaluating the response of macrophages and cancer cells in a hypoxic gradient.

bioengineering↗

bioRxiv2023

The impacts of active and self-supervised learning on efficient annotation of single-cell expression data

A crucial step in the analysis of single-cell data is annotating cells to cell types and states. While a myriad of approaches have been proposed, manual labeling of cells to create training datasets remains tedious and time-consuming. In the field of machine learning, active and self-supervised learning methods have been proposed to improve the performance of a classifier while reducing both annotation time and label budget. However, the benefits of such strategies for single-cell annotation have yet to be evaluated in realistic settings. Here, we perform a comprehensive benchmarking of active and self-supervised labeling strategies across a range of single-cell technologies and cell type annotation algorithms. We quantify the benefits of active learning and self-supervised strategies in the presence of cell type imbalance and variable similarity. We introduce several heuristic procedures tailored to single-cell data -- including a marker-aware adaptive reweighting cell selection procedure -- that shows competitive performance with existing approaches. In addition, we demonstrate that having prior knowledge of cell type markers improves annotation accuracy. Finally, we summarize our findings into a set of recommendations for those implementing cell type annotation procedures or platforms. An R package implementing the heuristic approaches introduced in this work may be found at https://github.com/camlab-bioml/leader.

bioinformatics↗

bioRxiv2022

The differential impacts of dataset imbalance in single-cell data integration

Single-cell transcriptomic data measured across distinct samples has led to a surge in computational methods for data integration. Few studies have explicitly examined the common case of cell-type imbalance between datasets to be integrated, and none have characterized its impact on downstream analyses. To address this gap, we developed the Iniquitate pipeline for assessing the stability of single-cell RNA sequencing (scRNA-seq) integration results after perturbing the degree of imbalance between datasets. Through benchmarking 5 state-of-the-art scRNA-seq integration techniques in 1600 perturbed integration scenarios for a multi-sample peripheral blood mononuclear cell (PBMC) dataset, our results indicate that sample imbalance has significant impacts on downstream analyses and the biological interpretation of integration results. We observed significant variation in clustering, cell-type classification, marker gene-based annotation, and query-to-reference mapping in imbalanced settings. Two key factors were found to lead to quantitation differences after scRNA-seq integration - the cell-type imbalance within and between samples (relative cell-type support) and the relatedness of cell-types across samples (minimum cell-type center distance). To account for evaluation gaps in imbalanced contexts, we developed novel clustering metrics robust to sample imbalance, including the balanced Adjusted Rand Index (bARI) and balanced Adjusted Mutual Information (bAMI). Our analysis quantifies biologically-relevant effects of dataset imbalance in integration scenarios and introduces guidelines and novel metrics for integration of disparate datasets. The Iniquitate pipeline and balanced clustering metrics are available at https://github.com/hsmaan/Iniquitate and https://github.com/hsmaan/balanced-clustering, respectively.

bioinformatics↗

bioRxiv2022

Multi-objective Bayesian Optimization with Heuristic Objectives for Biomedical and Molecular Data Analysis Workflows

Many practical applications require optimization of multiple, computationally expensive, and possibly competing objectives that are well-suited for multi-objective Bayesian optimization (MOBO) procedures. However, for many types of biomedical data, measures of data analysis workflow success are often heuristic and therefore it is not known a priori which objectives are useful. Thus, MOBO methods that return the full Pareto front may be suboptimal in these cases. Here we propose a novel MOBO method that adaptively updates the scalarization function using properties of the posterior of a multi-output Gaussian process surrogate function. This approach selects useful objectives based on a flexible set of desirable criteria, allowing the functional form of each objective to guide optimization. We demonstrate the qualitative behaviour of our method on toy data and perform proof-of-concept analyses of single-cell RNA sequencing and highly multiplexed imaging datasets.

bioinformatics↗

bioRxiv2021

Automated assignment of cell identity from single-cell multiplexed imaging and proteomic data

The creation of scalable single-cell and highly-multiplexed imaging technologies that profile the protein expression and phosphorylation status of heterogeneous cellular populations has led to multiple insights into disease processes including cancer initiation and progression. A major analytical challenge in interpreting the resulting data is the assignment of cells to a priori known cell types in a robust and interpretable manner. Existing approaches typically solve this by clustering cells followed by manual annotation of individual clusters or by strategies that gate protein expression at predefined thresholds. However, these often require several subjective analysis choices such as selecting the number of clusters and do not automatically assign cell types in line with prior biological knowledge. They further lack the ability to explicitly assign cells to an unknown or uncharacterized type, which exist in most highly multiplexed imaging experiments due to the limited number of markers quantified. To address these issues we present Astir, a probabilistic model to assign cells to cell types by integrating prior knowledge of marker proteins. Astir uses deep recognition neural networks for fast Bayesian inference, allowing for cell type annotations at the million-cell scale and in the absence of previously annotated reference data across multiple experimental modalities and antibody panels. We demonstrate that Astir outperforms existing approaches in terms of accuracy and robustness by applying it to over 2.1 million single cells from several suspension and imaging mass cytometry and microscopy datasets in multiple tissue contexts. We further showcase that Astir can be used for the fast analysis of the spatial architecture of the tumour microenvironment, automatically quantifying the immune influx and spatial heterogeneity of patient samples. Astir is freely available as an open source Python package at https://www.github.com/camlab-bioml/astir.

bioinformatics↗

bioRxiv2020

Single cell fitness landscapes induced by genetic and pharmacologic perturbations in cancer

Tumour fitness landscapes underpin selection in cancer, impacting etiology, evolution and response to treatment. Progress in defining fitness landscapes has been impeded by a lack of timeseries perturbation experiments over realistic intervals at single cell resolution. We studied the nature of clonal dynamics induced by genetic and pharmacologic perturbation with a quantitative fitness model developed to ascribe quantitative selective coefficients to individual cancer clones, enable prediction of clone-specific growth potential, and forecast competitive clonal dynamics over time. We applied the model to serial single cell genome (>60,000 cells) and transcriptome (>58,000 cells) experiments ranging from 10 months to 2.5 years in duration. We found that genetic perturbation of TP53 in epithelial cell lines induces multiple forms of copy number alteration that confer increased fitness to clonal populations with measurable consequences on gene expression. In patient derived xenografts, predicted selective coefficients accurately forecasted clonal competition dynamics, that were validated with timeseries sampling of experimentally engineered mixtures of low and high fitness clones. In cisplatin-treated patient derived xenografts, the fitness landscape was inverted in a time-dependent manner, whereby a drug resistant clone emerged from a phylogenetic lineage of low fitness clones, and high fitness clones were eradicated. Moreover, clonal selection mediated reversible drug response early in the selection process, whereas late dynamics in genomically fixed clones were associated with transcriptional plasticity on a fixed clonal genotype. Together, our findings outline causal mechanisms with implication for interpreting how mutations and multi-faceted drug resistance mechanisms shape the etiology and cellular fitness of human cancers.

cancer biology↗