arXiv subjects

Katherine A. Hoadley

Publications and source records attributed to Katherine A. Hoadley.

10 recordsLinked to original sources

arXiv2026

Mitigating Batch Effects in Histopathology via Language-Mediated Robust Embedding Generation

Pathology foundation models (PFMs) have demonstrated strong potential across clinical and scientific applications, yet their performance is often hindered by batch effects, which are non-biological variations across tissue source institutions (TSIs) that distort learned feature representations and impair generalization. Conventional mitigation strategies, such as stain normalization, offer limited success in addressing these high-dimensional, complex artifacts. We present GLMP (General-purpose LLM-Mediated Pathology model), a novel framework that generates robust numerical embeddings from histology image patches through an intermediate textual representation. By leveraging pretrained general-purpose multimodal large language models (MLLMs) and text encoders, GLMP effectively prioritizes biologically meaningful signals over TSI-specific artifacts, thereby improving cross-institutional generalization. To our knowledge, GLMP is the first pathology model to use text descriptions of histological features as an intermediate representation for generating numerical embeddings from histology images. Our results highlight the untapped potential of broad-domain, non-specialized MLLMs in computational pathology and introduce a new paradigm for building versatile, generalizable, and robust pathology models.

cs.CV↗

arXiv2025

Towards interpretable prediction of recurrence risk in breast cancer using pathology foundation models

Transcriptomic assays such as the PAM50-based ROR-P score guide recurrence risk stratification in non-metastatic, ER-positive, HER2-negative breast cancer but are not universally accessible. Histopathology is routinely available and may offer a scalable alternative. We introduce MAKO, a benchmarking framework evaluating 12 pathology foundation models and two non-pathology baselines for predicting ROR-P scores from H&E-stained whole slide images using attention-based multiple instance learning. Models were trained and validated on the Carolina Breast Cancer Study and externally tested on TCGA BRCA. Several foundation models outperformed baselines across classification, regression, and survival tasks. CONCH achieved the highest ROC AUC, while H-optimus-0 and Virchow2 showed top correlation with continuous ROR-P scores. All pathology models stratified CBCS participants by recurrence similarly to transcriptomic ROR-P. Tumor regions were necessary and sufficient for high-risk predictions, and we identified candidate tissue biomarkers of recurrence. These results highlight the promise of interpretable, histology-based risk models in precision oncology.

q-bio.TO↗

arXiv2023

Measure of Strength of Evidence for Visually Observed Differences between Subpopulations

For measuring the strength of visually-observed subpopulation differences, the Population Difference Criterion is proposed to assess the statistical significance of visually observed subpopulation differences. It addresses the following challenges: in high-dimensional contexts, distributional models can be dubious; in high-signal contexts, conventional permutation tests give poor pairwise comparisons. We also make two other contributions: Based on a careful analysis we find that a balanced permutation approach is more powerful in high-signal contexts than conventional permutations. Another contribution is the quantification of uncertainty due to permutation variation via a bootstrap confidence interval. The practical usefulness of these ideas is illustrated in the comparison of subpopulations of modern cancer data.

stat.ME↗

arXiv2022

Pairwise Nonlinear Dependence Analysis of Genomic Data

In The Cancer Genome Atlas (TCGA) data set, there are many interesting nonlinear dependencies between pairs of genes that reveal important relationships and subtypes of cancer. Such genomic data analysis requires a rapid, powerful and interpretable detection process, especially in a high-dimensional environment. We study the nonlinear patterns among the expression of pairs of genes from TCGA using a powerful tool called Binary Expansion Testing. We find many nonlinear patterns, some of which are driven by known cancer subtypes, some of which are novel.

stat.AP↗

arXiv2022

Jackstraw Inference for AJIVE Data Integration

In the age of big data, data integration is a critical step especially in the understanding of how diverse data types work together and work separately. Among data integration methods, the Angle-Based Joint and Individual Variation Explained (AJIVE) approach is particularly attractive because it not only studies joint behavior but also individual behavior. Typically AJIVE scores indicate important relationships between data objects, such as clusters. An important challenge is understanding which features, i.e. variables, are associated with those relationships. This challenge is addressed by the proposal of a hypothesis test for assessing statistical significance of features. The new test is inspired by the related jackstraw method developed for Principal Component Analysis. We use a high-dimensional muti-genomic cancer data set as our strong motivation and deep illustration of the methodology.

stat.AP↗

arXiv2022

Bidimensional linked matrix factorization for pan-omics pan-cancer analysis

Several modern applications require the integration of multiple large data matrices that have shared rows and/or columns. For example, cancer studies that integrate multiple omics platforms across multiple types of cancer, pan-omics pan-cancer analysis, have extended our knowledge of molecular heterogenity beyond what was observed in single tumor and single platform studies. However, these studies have been limited by available statistical methodology. We propose a flexible approach to the simultaneous factorization and decomposition of variation across such bidimensionally linked matrices, BIDIFAC+. This decomposes variation into a series of low-rank components that may be shared across any number of row sets (e.g., omics platforms) or column sets (e.g., cancer types). This builds on a growing literature for the factorization and decomposition of linked matrices, which has primarily focused on multiple matrices that are linked in one dimension (rows or columns) only. Our objective function extends nuclear norm penalization, is motivated by random matrix theory, gives an identifiable decomposition under relatively mild conditions, and can be shown to give the mode of a Bayesian posterior distribution. We apply BIDIFAC+ to pan-omics pan-cancer data from TCGA, identifying shared and specific modes of variability across 4 different omics platforms and 29 different cancer types.

stat.ML↗

arXiv2021

A Hierarchical Spike-and-Slab Model for Pan-Cancer Survival Using Pan-Omic Data

Pan-omics, pan-cancer analysis has advanced our understanding of the molecular heterogeneity of cancer, expanding what was known from single-cancer or single-omics studies. However, pan-cancer, pan-omics analyses have been limited in their ability to use information from multiple sources of data (e.g., omics platforms) and multiple sample sets (e.g., cancer types) to predict important clinical outcomes, like overall survival. We address the issue of prediction across multiple high-dimensional sources of data and multiple sample sets by using exploratory results from BIDIFAC+, a method for integrative dimension reduction of bidimensionally-linked matrices, in a predictive model. We apply a Bayesian hierarchical model that performs variable selection using spike-and-slab priors which are modified to allow for the borrowing of information across clustered data. This method is used to predict overall patient survival from the Cancer Genome Atlas (TCGA) using data from 29 cancer types and 4 omics sources. Our model selected patterns of variation identified by BIDIFAC+ that differentiate clinical tumor subtypes with markedly different survival outcomes. We also use simulations to evaluate the performance of the modified spike-and-slab prior in terms of its variable selection accuracy and prediction accuracy under different underlying data-generating frameworks. Software and code used for our analysis can be found at https://github.com/sarahsamorodnitsky/HierarchicalSS_PanCanPanOmics/ .

stat.AP↗

arXiv2020

Joint and individual analysis of breast cancer histologic images and genomic covariates

A key challenge in modern data analysis is understanding connections between complex and differing modalities of data. For example, two of the main approaches to the study of breast cancer are histopathology (analyzing visual characteristics of tumors) and genetics. While histopathology is the gold standard for diagnostics and there have been many recent breakthroughs in genetics, there is little overlap between these two fields. We aim to bridge this gap by developing methods based on Angle-based Joint and Individual Variation Explained (AJIVE) to directly explore similarities and differences between these two modalities. Our approach exploits Convolutional Neural Networks (CNNs) as a powerful, automatic method for image feature extraction to address some of the challenges presented by statistical analysis of histopathology image data. CNNs raise issues of interpretability that we address by developing novel methods to explore visual modes of variation captured by statistical algorithms (e.g. PCA or AJIVE) applied to CNN features. Our results provide many interpretable connections and contrasts between histopathology and genetics.

q-bio.QM↗

arXiv2019

A Pan-Cancer and Polygenic Bayesian Hierarchical Model for the Effect of Somatic Mutations on Survival

We built a novel Bayesian hierarchical survival model based on the somatic mutation profile of patients across 50 genes and 27 cancer types. The pan-cancer quality allows for the model to "borrow" information across cancer types, motivated by the assumption that similar mutation profiles may have similar (but not necessarily identical) effects on survival across different tissues-of-origin or tumor types. The effect of a mutation at each gene was allowed to vary by cancer type while the mean effect of each gene was shared across cancers. Within this framework we considered four parametric survival models (normal, log-normal, exponential, and Weibull), and we compared their performance via a cross-validation approach in which we fit each model on training data and estimate the log-posterior predictive likelihood on test data. The log-normal model gave the best fit, and we investigated the partial effect of each gene on survival via a forward selection procedure. Through this we determined that mutations at TP53 and FAT4 were together the most useful for predicting patient survival. We validated the model via simulation to ensure that our algorithm for posterior computation gave nominal coverage rates. The code used for this analysis can be found at http://github.com/sarahsamorodnitsky/Pan-Cancer-Survival-Modeling , and the results are at http://ericfrazerlock.com/surv_figs/SurvivalDisplay.html .

q-bio.QM↗

arXiv2013

Joint and individual variation explained (JIVE) for integrated analysis of multiple data types

Research in several fields now requires the analysis of data sets in which multiple high-dimensional types of data are available for a common set of objects. In particular, The Cancer Genome Atlas (TCGA) includes data from several diverse genomic technologies on the same cancerous tumor samples. In this paper we introduce Joint and Individual Variation Explained (JIVE), a general decomposition of variation for the integrated analysis of such data sets. The decomposition consists of three terms: a low-rank approximation capturing joint variation across data types, low-rank approximations for structured variation individual to each data type, and residual noise. JIVE quantifies the amount of joint variation between data types, reduces the dimensionality of the data and provides new directions for the visual exploration of joint and individual structures. The proposed method represents an extension of Principal Component Analysis and has clear advantages over popular two-block methods such as Canonical Correlation Analysis and Partial Least Squares. A JIVE analysis of gene expression and miRNA data on Glioblastoma Multiforme tumor samples reveals gene-miRNA associations and provides better characterization of tumor types. Data and software are available at https://genome.unc.edu/jive/

stat.ML↗