Ellis Patrick

dioscRi enables transferable prediction of clinical outcomes in multi-parameter cytometry data

dioscRi improves transferability of deep learning models for cytometry data by integrating normalization, cell type hierarchies, and clinical outcome prediction.

PACE, Proximity-Associated Changes in Expression

PACE quantifies how proximity to different cell types affects gene expression in spatial transcriptomics data, revealing tumour-associated reprogramming in breast and melanoma tissues.

Deep interpretable learning of sample representations for characterizing disease states in single-cell transcriptomics

This paper introduces Phenoverse, a deep learning framework that learns disease state representations from single-cell transcriptomic data, enabling disease prediction and severity assessment using binary labels.

Generalisable tissue-wide molecular reconstruction from histology

GHIST+ reconstructs tissue-wide molecular maps from H&E histology, enabling cohort-scale molecular reconstruction from sparse spatial transcriptomic data.

AI-Driven Spatial Transcriptomics Unlocks Large-Scale Breast Cancer Biomarker Discovery from Histopathology

Path2Space predicts spatial gene expression and tumor microenvironment from histopathology slides, identifying new breast cancer subgroups and improving biomarker prediction.

Interpretable deep generative ensemble learning for single-cell omics with Hydra

Hydra is a deep generative framework for single-cell omics data that improves cell type annotation and feature selection, particularly for rare populations.

Combining xenium in situ spatial transcriptomics and imaging mass cytometry on a single tissue section

This paper combines Xenium spatial transcriptomics and imaging mass cytometry on a single tissue section to analyze mRNA and protein at single-cell resolution, showing improved biological insights when integrating both data types.

Orchestrating Spatial Transcriptomics Analysis with Bioconductor

The paper provides an open-source online book for analyzing spatial transcriptomics data using Bioconductor in R, with reproducible code and datasets.

Spatial gene expression at single-cell resolution from histology using deep learning with GHIST

GHIST predicts spatial gene expression at single-cell resolution from histology images, showing superior performance across different resolutions and datasets.

SpatioMark: Quantifying the impact of spatial proximity on cell phenotype

SpatioMark quantifies how spatial proximity affects cell phenotype by analyzing spatial proteomics and transcriptomics data, linking findings to patient survival.

A highly adaptable protocol for mapping spatial features of cellular aggregates in tissues

This paper introduces a protocol to analyze spatial cellular structures in tissue lesions using two novel metrics and open-source software.

Tissue-wide profiling of human lungs reveals spatial sequestration of macrophages in tuberculosis

Xiao et al. reveal spatial segregation of immune cells in TB-lung tissue and link the microenvironmental dynamics to disease states of tuberculosis.

Kontextual: Reframing analysis of spatial omics data reveals consistent cell relationships across images

Spatial mapping reveals unique cellular interactions and enhanced tertiary lymphoid structures in responders to anti-PD-1 therapy in mucosal head and neck cancers.

The paper identifies immune landscape differences in responders and non-responders to anti-PD-1 therapy in head and neck cancers using spatial mapping.

Construction and optimization of multi-platform precision pathways for precision medicine

The paper develops a framework to construct multi-platform precision diagnostic pathways automatically, optimizing for accuracy and cost-effectiveness in clinical settings.

Benchmarking the translational potential of spatial gene expression prediction from histology

This study benchmarks six methods for predicting spatial gene expression from histology images using breast and skin cancer datasets and TCGA data, evaluating their performance and translational potential.

Atlas-scale single-cell multi-sample multi-condition data integration using scMerge2

scMerge2 integrates large-scale multi-sample multi-condition single-cell data across cohorts, improving disease progression discrimination through cell-type expression signatures.

Biologically-informed self-supervised learning for segmentation of subcellular spatial transcriptomics data

BIDCell improves subcellular spatial transcriptomics segmentation by combining self-supervised learning with biologically informed loss functions, outperforming existing methods across multiple tissue types.

MoleculeExperiment enables consistent infrastructure for molecule-resolved spatial transcriptomics data in Bioconductor

The paper introduces MoleculeExperiment, an R/Bioconductor package for handling molecule-resolved spatial transcriptomics data, enabling standardization and integration with existing single-cell tools.

Spatial analysis for highly multiplexed imaging data to identify tissue microenvironments

The paper introduces a statistical method to analyze highly multiplexed imaging data and identifies distinct tissue architectures across three high-parameter assays.

Deep Multimodal Graph-Based Network for Survival Prediction from Highly Multiplexed Images and Patient Variables

This paper introduces a deep multimodal graph-based network to predict breast cancer survival by integrating spatial phenotype data from imaging mass cytometry with patient variables.

The impact of similarity metrics on cell type clustering in highly multiplexed in situ imaging cytometry data

This study evaluates similarity metrics for clustering in situ imaging cytometry data and proposes FuseSOM, an ensemble algorithm that improves clustering performance.

scFeatures: Multi-view representations of single-cell and spatial data for disease outcome prediction

scFeatures creates interpretable molecular representations of single-cell and spatial data to predict disease outcomes using diverse feature types.

Cross-Platform Omics Prediction procedure enables precision medicine in patients with stage-III melanoma

The paper introduces a method to handle omics data variation across platforms and times, showing its use in three disease risk models for clinical application.

spicyR: Spatial analysis of in situ cytometry data in R

spicyR is an R package for analyzing spatial localization changes of cell types across images, applied to type 1 diabetes data to reveal disease-related cellular associations.

Identifying cellular-to-phenotype associations by elucidating hierarchical relationships in high-dimensional cytometry data

The paper introduces treekoR, a framework that recaptures hierarchical structures in cytometry data to better associate cell proportions with disease.

AFid: A tool for automated identification and exclusion of autofluorescent objects from microscopy images

AFid identifies and removes autofluorescent signals from multi-channel microscopy images to improve protein expression analysis.

Deconvolving the contributions of cell-type heterogeneity on cortical gene expression

The paper evaluates methods for estimating cell-type proportions in brain tissue using paired transcriptomic and immunohistochemistry data, showing some algorithms provide accurate estimates.

Investigating higher order interactions in single cell data with scHOT

The paper introduces scHOT, a method to detect changes in higher order gene interactions in single-cell data, demonstrating its application in liver development and spatial gene correlation patterns.

Evaluating stably expressed genes in single cells

This study evaluates stably expressed genes (SEGs) in single cells using a computational framework and finds they are more stable than traditional housekeeping genes across species.