TwinPurify: Purifying gene expression data to reveal tumor-intrinsic transcriptional programs via self-supervised learning

2026-01-26

AI Summary

Tumor purity variation can obscure intrinsic transcriptional signals and hinder downstream analyses of bulk transcriptomic data. To address this, this work proposes a reference-free, self-supervised representation learning framework that, for the first time, adapts the Barlow Twins objective to transcriptomic deconvolution. By leveraging adjacent normal tissues from the same cohort as contextual background, the method learns continuous, high-dimensional tumor embeddings that effectively disentangle tumor-specific signals. Departing from conventional deconvolution paradigms, the approach significantly outperforms baseline models such as autoencoders across multiple large-scale cancer cohorts. It not only enhances the accuracy of molecular subtyping and tumor grading, as well as the concordance of survival models, but also uncovers pathway activities with greater biological relevance.

Technology Category

• Machine Learning: Representation Learning
• Computer Vision: Representation Learning for Vision
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Advances in single-cell and spatial transcriptomic technologies have transformed tumor ecosystem profiling at cellular resolution. However, large scale studies on patient cohorts continue to rely on bulk transcriptomic data, where variation in tumor purity obscures tumor-intrinsic transcriptional signals and constrains downstream discovery. Many deconvolution methods report strong performance on synthetic bulk mixtures but fail to generalize to real patient cohorts because of unmodeled biological and technical variation. Here, we introduce TwinPurify, a representation learning framework that adapts the Barlow Twins self-supervised objective, representing a fundamental departure from the deconvolution paradigm. Rather than resolving the bulk mixture into discrete cell-type fractions, TwinPurify instead learns continuous, high-dimensional tumor embeddings by leveraging adjacent-normal profiles within the same cohort as"background"guidance, enabling the disentanglement of tumor-specific signals without relying on any external reference. Benchmarked against multiple large cancer cohorts across RNA-seq and microarray platforms, TwinPurify outperforms conventional representation learning baselines like auto-encoders in recovering tumor-intrinsic and immune signals. The purified embeddings improve molecular subtype and grade classification, enhance survival model concordance, and uncover biologically meaningful pathway activities compared to raw bulk profiles. By providing a transferable framework for decontaminating bulk transcriptomics, TwinPurify extends the utility of existing clinical datasets for molecular discovery.

Problem

• tumor purity
• bulk transcriptomics
• transcriptional programs
• deconvolution
• tumor-intrinsic signals

Innovation

• self-supervised learning
• tumor purity
• bulk transcriptomics
• representation learning
• Barlow Twins