Biology subjects

Taylor-Weiner, A.

Publications and source records attributed to Taylor-Weiner, A..

5 recordsLinked to original sources

bioRxiv2024

Spatial mapping of immunosuppressive cancer-associated fibroblast gene signatures in H&E-stained images using additive multiple instance learning

The relative abundance of cancer-associated fibroblast (CAF) subtypes influences a tumors response to treatment, especially immunotherapy. However, the extent to which the underlying tumor composition associates with CAF subtype-specific gene expression is unclear. Here, we describe an interpretable machine learning (ML) approach, additive multiple instance learning (aMIL), to predict bulk gene expression signatures from H&E-stained whole slide images (WSI), focusing on an immunosuppressive LRRC15+ CAF-enriched TGF{beta}-CAF signature. aMIL models accurately predicted TGF{beta}-CAF across various cancer types. Tissue regions contributing most highly to slide-level predictions of TGF{beta}-CAF were evaluated by ML models characterizing spatial distributions of diverse cell and tissue types, stromal subtypes, and nuclear morphology. In breast cancer, regions contributing most to TGF{beta}-CAF-high predictions ("excitatory") were localized to cancer stroma with high fibroblast density and mature collagen fibers. Regions contributing most to TGF{beta}-CAF-low predictions ("inhibitory") were localized to cancer epithelium and densely inflamed stroma. Fibroblast and lymphocyte nuclear morphology also differed between excitatory and inhibitory regions. Thus, aMIL enables a data-driven link between tissue phenotype and transcription, offering biological interpretability beyond typical black-box models.

cancer biology↗

bioRxiv2023

Cell-type-specific nuclear morphology predicts genomic instability and prognosis in multiple cancer types

While alterations in nucleus size, shape, and color are ubiquitous in cancer, comprehensive quantification of nuclear morphology across a whole-slide histologic image remains a challenge. Here, we describe the development of a pan-tissue, deep learning-based digital pathology pipeline for exhaustive nucleus detection, segmentation, and classification and the utility of this pipeline for nuclear morphologic biomarker discovery. Manually-collected nucleus annotations were used to train an object detection and segmentation model for identifying nuclei, which was deployed to segment nuclei in H&E-stained slides from the BRCA, LUAD, and PRAD TCGA cohorts. Interpretable features describing the shape, size, color, and texture of each nucleus were extracted from segmented nuclei and compared to measurements of genomic instability, gene expression, and prognosis. The nuclear segmentation and classification model trained herein performed comparably to previously reported models. Features extracted from the model revealed differences sufficient to distinguish between BRCA, LUAD, and PRAD. Furthermore, cancer cell nuclear area was associated with increased aneuploidy score and homologous recombination deficiency. In BRCA, increased fibroblast nuclear area was indicative of poor progression-free and overall survival and was associated with gene expression signatures related to extracellular matrix remodeling and anti-tumor immunity. Thus, we developed a powerful pan-tissue approach for nucleus segmentation and featurization, enabling the construction of predictive models and the identification of features linking nuclear morphology with clinically-relevant prognostic biomarkers across multiple cancer types.

cancer biology↗

bioRxiv2020

Dense, high-resolution mapping of cells and tissues from pathology images for the interpretable prediction of molecular phenotypes in cancer

While computational methods have made substantial progress in improving the accuracy and throughput of pathology workflows for diagnostic, prognostic, and genomic prediction, lack of interpretability remains a significant barrier to clinical integration. In this study, we present a novel approach for predicting clinically-relevant molecular phenotypes from histopathology whole-slide images (WSIs) using human-interpretable image features (HIFs). Our method leverages >1.6 million annotations from board-certified pathologists across >5,700 WSIs to train deep learning models for high-resolution tissue classification and cell detection across entire WSIs in five cancer types. Combining cell- and tissue-type models enables computation of 607 HIFs that comprehensively capture specific and biologically-relevant characteristics of multiple tumors. We demonstrate that these HIFs correlate with well-known markers of the tumor microenvironment (TME) and can predict diverse molecular signatures, including immune checkpoint protein expression and homologous recombination deficiency (HRD). Our HIF-based approach provides a novel, quantitative, and interpretable window into the composition and spatial architecture of the TME.

bioinformatics↗

bioRxiv2019

Passenger Hotspot Mutations in Cancer

Hotspots, or mutations that recur at the same genomic site across multiple tumors, have been conventionally interpreted as strong universal evidence of somatic positive selection, unequivocally pinpointing genes driving tumorigenesis. Here, we demonstrate that this convention is falsely premised on an inaccurate statistical model of background mutagenesis. Many hotspots are in fact passenger events, recurring at sites that are simply inherently more mutable rather than under positive selection, which current background models do not account for. We thus detail a log-normal-Poisson (LNP) background model that accounts for variation in site-specific mutability in a manner consistent with models of mutagenesis, use this model to show that the tendency to generate passenger hotspots pervades all common mutational processes, and apply it to a ~10, 000 patient cohort from The Cancer Genome Atlas to nominate driver hotspots with far fewer false positives compared to conventional methods. As the biomedical community faces critical decisions in prioritizing putative driver mutations for deep experimental characterization to assess therapeutic potential, we offer our findings as a guide to avoid wasting valuable scientific resources on passenger hotspots.

genomics↗

bioRxiv2018

Discovery of cancer driver genes based on nucleotide context

Many cancer genomes contain large numbers of somatic mutations, but few of these mutations drive tumor development. Current approaches to identify cancer driver genes are largely based on mutational recurrence, i.e. they search for genes with an increased number of nonsynonymous mutations relative to the local background mutation rate. Multiple studies have noted that the sensitivity of recurrence-based methods is limited in tumors with high background mutation rates, because passenger mutations dilute their statistical power. Here, we observe that passenger mutations tend to occur in characteristic nucleotide sequence contexts, while driver mutations follow a different distribution pattern determined by the location of functionally relevant genomic positions along the protein-coding sequence. To discover new cancer genes, we searched for genes with an excess of mutations in unusual nucleotide contexts that deviate from the characteristic context around passenger mutations. By applying this statistical framework to whole-exome sequencing data from 12,004 tumors, we discovered a long tail of novel candidate cancer genes with mutation frequencies as low as 1% and functional supporting evidence. Our results show that considering both the number and the nucleotide context around mutations helps identify novel cancer driver genes, particularly in tumors with high background mutation rates.

genetics↗