Biologically Disentangled Multi-Omic Modeling Reveals Mechanistic Insights into Pan-Cancer Immunotherapy Resistance

• URL: http://arxiv.org/abs/2508.18638v1
• Date: Tue, 26 Aug 2025 03:33:56 GMT
• Title: Biologically Disentangled Multi-Omic Modeling Reveals Mechanistic Insights into Pan-Cancer Immunotherapy Resistance
• Authors: Ifrah Tariq, Ernest Fraenkel,
• Abstract summary: We introduce the Biologically Disentangled Variational Autoencoder (BDVAE), a deep generative model that integrates transcriptomic and genomic data.<n>Applying to a pan-cancer cohort of 366 patients, BDVAE accurately predicts treatment response.<n>It uncovers critical resistance mechanisms, including immune suppression, metabolic shifts, and neuronal signaling.
• Score: 0.9787436863401008
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Immune checkpoint inhibitors (ICIs) have transformed cancer treatment, yet patient responses remain highly variable, and the biological mechanisms underlying resistance are poorly understood. While machine learning models hold promise for predicting responses to ICIs, most existing methods lack interpretability and do not effectively leverage the biological structure inherent to multi-omics data. Here, we introduce the Biologically Disentangled Variational Autoencoder (BDVAE), a deep generative model that integrates transcriptomic and genomic data through modality- and pathway-specific encoders. Unlike existing rigid, pathway-informed models, BDVAE employs a modular encoder architecture combined with variational inference to learn biologically meaningful latent features associated with immune, genomic, and metabolic processes. Applied to a pan-cancer cohort of 366 patients across four cancer types treated with ICIs, BDVAE accurately predicts treatment response (AUC-ROC = 0.94 on unseen test data) and uncovers critical resistance mechanisms, including immune suppression, metabolic shifts, and neuronal signaling. Importantly, BDVAE reveals that resistance spans a continuous biological spectrum rather than strictly binary states, reflecting gradations of tumor dysfunction. Several latent features correlate with survival outcomes and known clinical subtypes, demonstrating BDVAE's capability to generate interpretable, clinically relevant insights. These findings underscore the value of biologically structured machine learning in elucidating complex resistance patterns and guiding precision immunotherapy strategies.

Related papers

• Contextual Invertible World Models: A Neuro-Symbolic Agentic Framework for Colorectal Cancer Drug Response [4.796382757669091]
  Precision oncology is currently limited by the small-N, large-P paradox.<n>We present a Neuro-Symbolic Agentic Framework that bridges this gap.<n>Our framework provides a transparent, biologically grounded path towards explainable AI in cancer research.
  arXiv  Detail & Related papers  (2026-03-01T16:15:58Z)
• Predicting Multi-Drug Resistance in Bacterial Isolates Through Performance Comparison and LIME-based Interpretation of Classification Models [0.0]
  This study proposes an interpretable machine learning framework to predict Multi-Drug Resistance (MDR) in bacterial isolates.<n>Five classification models were evaluated, including Logistic Regression, Random Forest, AdaBoost, XGBoost, and LightGBM.<n> Ensemble models, particularly XGBoost and LightGBM, demonstrated superior predictive capability across all metrics.
  arXiv  Detail & Related papers  (2026-02-25T20:50:01Z)
• DOGMA: Weaving Structural Information into Data-centric Single-cell Transcriptomics Analysis [43.565183518761984]
  We propose DOGMA, a data-centric framework designed for the structural reshaping and semantic enhancement of raw data.<n>In complex multi-species and multi-organ benchmarks, DOGMA SOTA performance, exhibiting superior zero-shot robustness and sample efficiency.
  arXiv  Detail & Related papers  (2026-02-02T09:10:09Z)
• Investigating the Impact of Histopathological Foundation Models on Regressive Prediction of Homologous Recombination Deficiency [52.50039435394964]
  We systematically evaluate foundation models for regression-based tasks.<n>We extract patch-level features from whole slide images (WSI) using five state-of-the-art foundation models.<n>Models are trained to predict continuous HRD scores based on these extracted features across breast, endometrial, and lung cancer cohorts.
  arXiv  Detail & Related papers  (2026-01-29T14:06:50Z)
• DISPROTBENCH: A Disorder-Aware, Task-Rich Benchmark for Evaluating Protein Structure Prediction in Realistic Biological Contexts [76.59606029593085]
  DisProtBench is a benchmark for evaluating protein structure prediction models (PSPMs) under structural disorder and complex biological conditions.<n>DisProtBench spans three key axes: data complexity, task diversity, and Interpretability.<n>Results reveal significant variability in model robustness under disorder, with low-confidence regions linked to functional prediction failures.
  arXiv  Detail & Related papers  (2025-06-18T23:58:22Z)
• Interpretable Graph Kolmogorov-Arnold Networks for Multi-Cancer Classification and Biomarker Identification using Multi-Omics Data [36.92842246372894]
  Multi-Omics Graph Kolmogorov-Arnold Network (MOGKAN) is a deep learning framework that utilizes messenger-RNA, micro-RNA sequences, and DNA methylation samples.<n>By integrating multi-omics data with graph-based deep learning, our proposed approach demonstrates robust predictive performance and interpretability.
  arXiv  Detail & Related papers  (2025-03-29T02:14:05Z)
• TopoTxR: A topology-guided deep convolutional network for breast parenchyma learning on DCE-MRIs [49.69047720285225]
  We propose a novel topological approach that explicitly extracts multi-scale topological structures to better approximate breast parenchymal structures. We empirically validate emphTopoTxR using the VICTRE phantom breast dataset. Our qualitative and quantitative analyses suggest differential topological behavior of breast tissue in treatment-na"ive imaging.
  arXiv  Detail & Related papers  (2024-11-05T19:35:10Z)
• LASSO-MOGAT: A Multi-Omics Graph Attention Framework for Cancer Classification [41.94295877935867]
  This paper introduces LASSO-MOGAT, a graph-based deep learning framework that integrates messenger RNA, microRNA, and DNA methylation data to classify 31 cancer types.
  arXiv  Detail & Related papers  (2024-08-30T16:26:04Z)
• Variational and Explanatory Neural Networks for Encoding Cancer Profiles and Predicting Drug Responses [40.80133767939435]
  Existing AI models face challenges due to noise in transcriptomics data and lack of biological interpretability. We introduce VETE, a novel neural network framework that incorporates a variational component to mitigate noise effects. VETE bridges the gap between AI-driven predictions and biologically meaningful insights in cancer research.
  arXiv  Detail & Related papers  (2024-07-05T13:13:02Z)
• Functional Integrative Bayesian Analysis of High-dimensional Multiplatform Genomic Data [0.8029049649310213]
  We propose a framework called Functional Integrative Bayesian Analysis of High-dimensional Multiplatform Genomic Data (fiBAG) fiBAG allows simultaneous identification of upstream functional evidence of proteogenomic biomarkers. We demonstrate the profitability of fiBAG via a pan-cancer analysis of 14 cancer types.
  arXiv  Detail & Related papers  (2022-12-29T03:31:45Z)
• Differentiable Agent-based Epidemiology [71.81552021144589]
  We introduce GradABM: a scalable, differentiable design for agent-based modeling that is amenable to gradient-based learning with automatic differentiation. GradABM can quickly simulate million-size populations in few seconds on commodity hardware, integrate with deep neural networks and ingest heterogeneous data sources.
  arXiv  Detail & Related papers  (2022-07-20T07:32:02Z)
• COVID-Net Biochem: An Explainability-driven Framework to Building Machine Learning Models for Predicting Survival and Kidney Injury of COVID-19 Patients from Clinical and Biochemistry Data [66.43957431843324]
  We introduce COVID-Net Biochem, a versatile and explainable framework for constructing machine learning models. We apply this framework to predict COVID-19 patient survival and the likelihood of developing Acute Kidney Injury during hospitalization.
  arXiv  Detail & Related papers  (2022-04-24T07:38:37Z)
• Improving generalization of machine learning-identified biomarkers with causal modeling: an investigation into immune receptor diagnostics [2.40246230430283]
  We focus on a specific, recently established high-dimensional biomarker - adaptive immune receptor repertoires (AIRRs) We argue that causal modeling improves machine learning-based biomarker robustness by identifying stable relations between variables.
  arXiv  Detail & Related papers  (2022-04-20T08:15:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.