Inference of single cell profiles from histology stains with the Single-Cell omics from Histology Analysis Framework (SCHAF)

Comiter, Charles; Vaishnav, Eeshit Dhaval; Ciapmricotti, Metamia; Li, Bo; Yang, Yiming; Rodig, Scott J.; Turner, Madison; Pfaff, Kathleen L.; Jané‐Valbuena, Judit; Slyper, Michal; Waldman, Julia; Vigneau, Sébastien; Wu, Jingyi; Blosser, Timothy R.; Segerstolpe, Åsa; Abravanel, Daniel L.; Wagle, Nikil; Zhuang, Xiaowei; Rudin, Charles M.; Klughammer, Johanna; Rozenblatt-Rosen, Orit; Kobayash-Kirschvink, Koseki J; Shu, Jian; Regev, Aviv

doi:10.1101/2023.03.21.533680

Cited by 20 publications

(10 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, methods predicting or incorporating scRNA-seq (e.g. SCHAF 45 , St2cell 46 and 47 ) or bulk RNA-seq (e.g. HE2RNA 48 and ISG 49 ) were excluded.…”

Section: Methodsmentioning

confidence: 99%

Benchmarking the translational potential of spatial gene expression prediction from histology

Chan,

Wang,

et al. 2023

Preprint

View full text Add to dashboard Cite

Spatial transcriptomics has enabled the quantification of gene expression at spatial coordinates, offering crucial insights into molecular underpinnings of diseases. In light of this, several methods predicting spatial gene expression from paired histology images have offered the opportunity of enhancing the utility of readily obtainable and cost-effective haematoxylin-and-eosin-stained histology images. To this end, we conducted a comprehensive benchmarking study encompassing six developed methods. These methods were reproduced and evaluated using HER2-positive breast tumour and human cutaneous squamous cell carcinoma datasets, followed by external validation using The Cancer Genome Atlas data. Our evaluation incorporates diverse metrics which capture the performance of predicted gene expression, model generalisability, translational potential, usability and computational efficiency of each method. Our findings demonstrate the capacity of methods to spatial gene expression from histology and highlight key areas that can be addressed to support the advancement of this emerging field.

show abstract

“…In addition, methods predicting or incorporating scRNA-seq (e.g. SCHAF 45 , St2cell 46 and 47 ) or bulk RNA-seq (e.g. HE2RNA 48 and ISG 49 ) were excluded.…”

Section: Methodsmentioning

confidence: 99%

Benchmarking the translational potential of spatial gene expression prediction from histology

Chan,

Wang,

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…This is seen from the variation across targets in our spatial co-expression examples. Comiter et al [67] recently developed ML models that produce spatially resolved 'omics data from H&E, using a training set of aligned H&E and scRNA-seq (versus bulk) data, and showed reasonable spatial correlation, at the tile level, with both pathologist scores and measured spatial expression. In ongoing work, we are collecting spatial expression data, which would allow a more quantitative assessment of our spatial predictions and even training models to directly predict measured spatial 'omics.…”

Section: Machine Learning Lessonsmentioning

confidence: 99%

Machine learning enabled prediction of digital biomarkers from whole slide histopathology images

McCaw,

Shcherbina,

Shah

et al. 2024

Preprint

View full text Add to dashboard Cite

Current predictive biomarkers generally leverage technologies such as immunohis-tochemistry or genetic analysis, which may require specialized equipment, be time-intensive to deploy, or incur human error. In this paper, we present an alternative approach for the development and deployment of a class of predictive biomarkers, leveraging deep learning on digital images of hematoxylin and eosin (H&E)-stained biopsy samples to simultaneously predict a range of molecular factors that are relevant to treatment selection and response. Our framework begins with the training of a pan-solid tumor H&E foundation model, which can generate a universal featurization of H&E-stained tissue images. This featurization becomes the input to machine learning models that perform multi-target, pan-cancer imputation. For a set of 352 drug targets, we show the ability to predict with high accuracy: copy number amplifications, target RNA expression, and an RNA-derived “amplification signature” that captures the transcriptional consequences of an amplification event. We facilitate exploratory analyses by making broad predictions initially. Having identified the subset of biomarkers relevant to a patient population of interest, we develop specialized machine learning models, built on the same foundational featurization, which achieve even higher performance for key biomarkers in tumor types of interest. Moreover, our models are robust, generalizing with minimal loss of performance across different patient populations. By generating imputations from tile-level featurizations, we enable spatial overlays of molecular annotations on top of whole-slide images. These annotation maps provide a clear means of interpreting the histological correlates of our model’s predictions, and align with features identified by expert pathologist review. Overall, our work demonstrates a flexible and scalable framework for imputing molecular measurements from H&E, providing a generalizable approach to the development and deployment of predictive biomarkers for targeted therapeutics in cancer.

show abstract

“…Although it is still early days for such approaches, the first projects following this strategy have already provided promising preliminary results (Irmisch et al, 2021). At the same time, machine learning models can predict spatially resolved single-cell profiles from H&E images, strengthening the link between the phenotype and omics profiles (preprint: Comiter et al, 2023). However, it remains to be seen which cellular programs, or intensities of genes, proteins, or metabolites can be predicted and how this information can be used in applications.…”

Section: Clinical and Beyondmentioning

confidence: 99%

Enablers and challenges of spatial omics, a melting pot of technologies

Alexandrov,

Saez‐Rodriguez,

Saka

2023

Molecular Systems Biology

View full text Add to dashboard Cite

Spatial omics has emerged as a rapidly growing and fruitful field with hundreds of publications presenting novel methods for obtaining spatially resolved information for any omics data type on spatial scales ranging from subcellular to organismal. From a technology development perspective, spatial omics is a highly interdisciplinary field that integrates imaging and omics, spatial and molecular analyses, sequencing and mass spectrometry, and image analysis and bioinformatics. The emergence of this field has not only opened a window into spatial biology, but also created multiple novel opportunities, questions, and challenges for method developers. Here, we provide the perspective of technology developers on what makes the spatial omics field unique. After providing a brief overview of the state of the art, we discuss technological enablers and challenges and present our vision about the future applications and impact of this melting pot.

show abstract

Inference of single cell profiles from histology stains with the Single-Cell omics from Histology Analysis Framework (SCHAF)

Cited by 20 publications

References 36 publications

Benchmarking the translational potential of spatial gene expression prediction from histology

Benchmarking the translational potential of spatial gene expression prediction from histology

Machine learning enabled prediction of digital biomarkers from whole slide histopathology images

Enablers and challenges of spatial omics, a melting pot of technologies

Contact Info

Product

Resources

About