Adversarial Domain Adaptation Enables Knowledge Transfer Across Heterogeneous RNA-Seq Datasets

Kevin Dradjat; Massinissa Hamidi; Blaise Hanczar

cs.LGMar 9, 2026

Adversarial Domain Adaptation Enables Knowledge Transfer Across Heterogeneous RNA-Seq Datasets

Authors: Kevin Dradjat, Massinissa Hamidi, Blaise Hanczar

Organizations: IBISC Laboratory, University Paris-Saclay (Univ. Evry), France · ADLIN, Paris, France

Abstract

Accurate phenotype prediction from RNA sequencing (RNA-seq) data is essential for diagnosis, biomarker discovery, and personalized medicine. Deep learning models have demonstrated strong potential to outperform classical machine learning approaches, but their performance relies on large, well-annotated datasets. In transcriptomics, such datasets are frequently limited, leading to over-fitting and poor generalization. Knowledge transfer from larger, more general datasets can alleviate this issue. However, transferring information across RNA-seq datasets remains challenging due to heterogeneous preprocessing pipelines and differences in target phenotypes. In this study, we propose a deep learning-based domain adaptation framework that enables effective knowledge transfer from a large general dataset to a smaller one for cancer type classification. The method learns a domain-invariant latent space by jointly optimizing classification and domain alignment objectives. To ensure stable training and robustness in data-scarce scenarios, the framework is trained with an adversarial approach with appropriate regularization. Both supervised and unsupervised approach variants are explored, leveraging labeled or unlabeled target samples. The framework is evaluated on three large-scale transcriptomic datasets (TCGA, ARCHS4, GTEx) to assess its ability to transfer knowledge across cohorts. Experimental results demonstrate consistent improvements in cancer and tissue type classification accuracy compared to non-adaptive baselines, particularly in low-data scenarios. Overall, this work highlights domain adaptation as a powerful strategy for data-efficient knowledge transfer in transcriptomics, enabling robust phenotype prediction under constrained data conditions.

Figures & tables

Figure 1: Conceptual illustration of domain adaptation process.

Figure 2: Composition of the domain adaptation pipeline.

E

C

and

D

denote, respectively, the encoder, the phenotype classifier and the domain discriminator. The unsupervised variant is defined by only the solid arrows, whereas we add the dotted arrows for the supervised variant. The discriminator loss can either be the Wasserstein distance or cross-entropy.

Figure 3: UMAP visualisations after domain adaptation for two study cases: (a) TCGA-target and (b) GTEx-target. Top row shows embeddings coloured by domain and bottom row shows embeddings coloured by class. (Top) TCGA dataset, (Bottom) GTEx dataset.

\multirow 3*Method	non-adaptive				adaptive
	Target-only	Supervised	Combat	limma	Wasserstein	Wasserstein	DANN	DANN
		no adaptation			unsupervised	supervised	unsupervised	supervised
TCGA	98.76 $\pm$ 0.1	91.91 $\pm$ 0.1	97.88 $\pm$ 0.086	97.56 $\pm$ 0.05	73.13 $\pm$ 0.72	97.18 $\pm$ 0.17	63.95 $\pm$ 1.82	96.94 $\pm$ 0.25
GTEx	98.93 $\pm$ 0.04	99.32 $\pm$ 0.07	98.25 $\pm$ 0.55	98.25 $\pm$ 0.53	22.38 $\pm$ 0.48	99.38 $\pm$ 0.04	51.12 $\pm$ 1.42	99.96 $\pm$ 0.04

Table 1: Accuracy on the TCGA/GTEx target test set after each adaptation method when considering the entire source and target training sets.

Figure 4: Performances of each adaptation methods by considering the TCGA dataset as target (a) and GTEx dataset as target (b).

Figure 5: Test target accuracy for different proportion of source-domain training used for (a) TCGA and (b) GTEx targets. Each model was trained with an increasing proportion of source samples while keeping labeled target proportion fixed to $0.01$ ( $\sim$ 100 examples).

Explore similar work

Mar 17, 2026

Sample-Efficient Adaptation of Drug-Response Models to Patient Tumors under Strong Biological Domain Shift

Predicting drug response in patients from preclinical data remains a major challenge in precision oncology due to the substantial biological gap between in vitro cell lines and patient tumors. Rather than aiming to improve absolute in vitro prediction accuracy, this work examines whether explicitly separating representation learning from task supervision enables more sample-efficient adaptation of drug-response models to patient data under strong biological domain shift. We propose a staged transfer-learning framework in which cellular and drug representations are first learned independently from large collections of unlabeled pharmacogenomic data using autoencoder-based representation learning. These representations are then aligned with drug-response labels on cell-line data and subsequently adapted to patient tumors using few-shot supervision. Through a systematic evaluation spanning in-domain, cross-dataset, and patient-level settings, we show that unsupervised pretraining provides limited benefit when source and target domains overlap substantially, but yields clear gains when adapting to patient tumors with very limited labeled data. In particular, the proposed framework achieves faster performance improvements during few-shot patient-level adaptation while maintaining comparable accuracy to single-phase baselines on standard cell-line benchmarks. Overall, these results demonstrate that learning structured and transferable representations from unlabeled molecular profiles can substantially reduce the amount of clinical supervision required for effective drug-response prediction, offering a practical pathway toward data-efficient preclinical-to-clinical translation.

Camille Jimenez Cortes, Philippe Lalanda, German Vega

Université Grenoble Alpes (UGA), France

Cross-Domain Transfer Learning Few-Shot Domain Adaptation Clinical Prediction Molecular Representation Learning

Ask PDF

Apr 2, 2026

Ranking-Guided Semi-Supervised Domain Adaptation for Severity Classification

Semi-supervised domain adaptation leverages a few labeled and many unlabeled target samples, making it promising for addressing domain shifts in medical image analysis. However, existing methods struggle with severity classification due to unclear class boundaries. Severity classification involves naturally ordered class labels, complicating adaptation. We propose a novel method that aligns source and target domains using rank scores learned via ranking with class order. Specifically, Cross-Domain Ranking ranks sample pairs across domains, while Continuous Distribution Alignment aligns rank score distributions. Experiments on ulcerative colitis and diabetic retinopathy classification validate the effectiveness of our approach, demonstrating successful alignment of class-specific rank score distributions.

Shota Harada, Ryoma Bise, Kiyohito Tanaka +1

Kyushu University, Japan · Kyoto Second Red Cross Hospital, Japan

Medical Image Classification Distribution Matching

Ask PDF

Jan 23, 2026

Semi-Supervised Domain Adaptation with Latent Diffusion for Pathology Image Classification

Deep learning models in computational pathology often fail to generalize across cohorts and institutions due to domain shift. Existing approaches either fail to leverage unlabeled data from the target domain or rely on image-to-image translation, which can distort tissue structures and compromise model accuracy. In this work, we propose a semi-supervised domain adaptation (SSDA) framework that utilizes a latent diffusion model trained on unlabeled data from both the source and target domains to generate morphology-preserving and target-aware synthetic images. By conditioning the diffusion model on foundation model features, cohort identity, and tissue preparation method, we preserve tissue structure in the source domain while introducing target-domain appearance characteristics. The target-aware synthetic images, combined with real, labeled images from the source cohort, are subsequently used to train a downstream classifier, which is then tested on the target cohort. The effectiveness of the proposed SSDA framework is demonstrated on the task of lung adenocarcinoma prognostication. The proposed augmentation yielded substantially better performance on the held-out test set from the target cohort, without degrading source-cohort performance. The approach improved the weighted F1 score on the target-cohort held-out test set from 0.611 to 0.706 and the macro F1 score from 0.641 to 0.716. Our results demonstrate that target-aware diffusion-based synthetic data augmentation provides a promising and effective approach for improving domain generalization in computational pathology.

Tengyue Zhang, Ruiwen Ding, Luoting Zhuang +3

Medical & Imaging Informatics, Department of Radiological Sciences, David Geffen School of Medicine, University of California, Los Angeles (UCLA), Los Angeles, CA 90024, USA · Bioengineering Department, Henry Samueli School of Engineering, UCLA, Los Angeles, CA 90024, USA · Department of Pathology & Laboratory Sciences, David Geffen School of Medicine, UCLA, Los Angeles, CA 90024, USA

Computational Pathology Synthetic Data Augmentation Latent Diffusion Models Histopathology Image Classification

Ask PDF