MINT: Molecularly Informed Training with Spatial Transcriptomics Supervision for Pathology Foundation Models

• URL: http://arxiv.org/abs/2603.07895v1
• Date: Mon, 09 Mar 2026 02:29:10 GMT
• Title: MINT: Molecularly Informed Training with Spatial Transcriptomics Supervision for Pathology Foundation Models
• Abstract summary: We propose MINT, a fine-tuning framework that incorporates spatial transcriptomics supervision into pretrained pathology Vision Transformers.<n> MINT appends a learnable ST token to the ViT input to encode transcriptomic information separately from the morphological CLS token.<n> MINT achieves the best overall performance on both HEST-Bench for gene expression prediction and general pathology tasks.
• Score: 5.15189458127282
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Pathology foundation models learn morphological representations through self-supervised pretraining on large-scale whole-slide images, yet they do not explicitly capture the underlying molecular state of the tissue. Spatial transcriptomics technologies bridge this gap by measuring gene expression in situ, offering a natural cross-modal supervisory signal. We propose MINT (Molecularly Informed Training), a fine-tuning framework that incorporates spatial transcriptomics supervision into pretrained pathology Vision Transformers. MINT appends a learnable ST token to the ViT input to encode transcriptomic information separately from the morphological CLS token, preventing catastrophic forgetting through DINO self-distillation and explicit feature anchoring to the frozen pretrained encoder. Gene expression regression at both spot-level (Visium) and patch-level (Xenium) resolutions provides complementary supervision across spatial scales. Trained on 577 publicly available HEST samples, MINT achieves the best overall performance on both HEST-Bench for gene expression prediction (mean Pearson r = 0.440) and EVA for general pathology tasks (0.803), demonstrating that spatial transcriptomics supervision complements morphology-centric self-supervised pretraining.

Related papers

• Towards Reliable Stain Transfer: An Iterative Data-Model Co-Optimization Framework Based on Multimodal Expert-Guided Assessment [55.16957949238588]
  We propose DMCoStain, a novel Data-Model Co-optimization framework for Stain transfer.<n>It iteratively co-refines training data and model capability, improving staining accuracy and interpretability in both pathological and structural consistency.<n>To support MEGFS, we construct ImmunoInstruction, the first large-scale IPE instruction-following dataset with 150K VQA samples.
  arXiv  Detail & Related papers  (2026-07-28T07:54:43Z)
• Spatial Transcriptomics-Guided Alignment Enhances Molecular Profiling in Pathology Foundation Model [43.24445461533517]
  We present a spatial transcriptomics-guided alignment framework for Molecular Profiling (STAMP)<n>STAMP applies a pathway-informed alignment strategy that aggregates transcriptomic data into biologically functional pathways.<n>The clinical utility of these augmented representations was validated through a multi-tier evaluation framework.
  arXiv  Detail & Related papers  (2026-05-29T16:41:14Z)
• Beyond ViT Tokens: Masked-Diffusion Pretrained Convolutional Pathology Foundation Model for Cell-Level Dense Prediction [33.957729129082075]
  Masked-Diffusion Convolutional Foundation Models (CMD) is a self-supervised convolutional generative pretraining framework for dense pathology representation learning.<n> CMD consistently outperforms existing ViT-based pathology foundation models and even surpasses state-of-the-art end-to-end segmentation methods.<n>Our findings suggest that purely convolutional architectures can also serve as competitive pathology foundation models for cell-level dense prediction.
  arXiv  Detail & Related papers  (2026-05-08T04:34:56Z)
• Learning to Synergize Semantic and Geometric Priors for Limited-Data Wheat Disease Segmentation [51.06355116973389]
  We propose a framework that treats wheat disease segmentation as a coupled task of disease-specific semantic perception and disease boundary localization.<n>Our core insight is that pretrained DINOv2 provides robust category-aware semantic priors to handle appearance shifts.<n>We show that SGPer consistently achieves state-of-the-art performance on wheat disease and organ segmentation benchmarks.
  arXiv  Detail & Related papers  (2026-04-07T04:19:39Z)
• Adapting a Pre-trained Single-Cell Foundation Model to Spatial Gene Expression Generation from Histology Images [5.072469295071516]
  HINGE (HIstology-coNditioned GEneration) retrofits a pre-trained sc-FM into a conditional expression generator.<n> Evaluated on three ST datasets, ours outperforms state-of-the-art baselines on mean Pearson correlation.
  arXiv  Detail & Related papers  (2026-03-20T08:52:39Z)
• Layout-Guided Controllable Pathology Image Generation with In-Context Diffusion Transformers [57.54843029965778]
  Controllable pathology image synthesis requires reliable regulation of spatial layout, tissue morphology, and semantic detail.<n>We propose In-Context Diffusion Transformer (IC-DiT), a layout-aware generative model that incorporates spatial layouts, textual descriptions, and visual embeddings into a unified diffusion transformer.<n>IC-DiT achieves higher fidelity, stronger spatial controllability, and better diagnostic consistency than existing methods.
  arXiv  Detail & Related papers  (2026-03-11T06:14:11Z)
• Towards Spatial Transcriptomics-driven Pathology Foundation Models [32.70436266943553]
  We introduce a vision-omics self-supervised learning framework that infuses localized molecular information into pathology vision encoders.<n>We instantiate SEAL by training on over 700,000 paired gene expression spot-tissue region examples spanning tumor and normal samples from 14 organs.<n> SEAL encoders exhibit robust domain generalization on out-of-distribution and enable new cross-modal capabilities such as gene-to-image retrieval.
  arXiv  Detail & Related papers  (2026-02-15T15:06:45Z)
• Fusing Pixels and Genes: Spatially-Aware Learning in Computational Pathology [46.83014413674925]
  STAMP is a spatial transcriptomics-augmented multimodal pathology representation learning framework.<n>Our study shows that self-supervised, gene-guided training provides a robust and task-agnostic signal for learning pathology image representations.<n>We validate STAMP across six datasets and four downstream tasks, where it consistently achieves strong performance.
  arXiv  Detail & Related papers  (2026-02-15T00:59:13Z)
• Supervise Less, See More: Training-free Nuclear Instance Segmentation with Prototype-Guided Prompting [53.799446807827714]
  We introduce SPROUT, a fully training- and annotation-free prompting framework for nuclear instance segmentation.<n> SPROUT leverages histology-informed priors to construct slide-specific reference prototypes.<n>The resulting foreground and background features are transformed into positive and negative point prompts, enabling the Segment Anything Model (SAM) to produce precise nuclear delineations.
  arXiv  Detail & Related papers  (2025-11-25T05:58:33Z)
• A Large-Scale Benchmark of Cross-Modal Learning for Histology and Gene Expression in Spatial Transcriptomics [8.854289521774483]
  HESCAPE is a benchmark for evaluating cross-modal contrastive pretraining in spatial transcriptomics.<n>Gene models pretrained on spatial transcriptomics data outperform both those trained without spatial data and simple baseline approaches.<n>We identify batch effects as a key factor that interferes with effective cross-modal alignment.
  arXiv  Detail & Related papers  (2025-08-02T21:11:36Z)
• MIRROR: Multi-Modal Pathological Self-Supervised Representation Learning via Modality Alignment and Retention [57.044719143401664]
  Histopathology and transcriptomics are fundamental modalities in oncology, encapsulating the morphological and molecular aspects of the disease.<n>We present MIRROR, a novel multi-modal representation learning method designed to foster both modality alignment and retention.<n>Extensive evaluations on TCGA cohorts for cancer subtyping and survival analysis highlight MIRROR's superior performance.
  arXiv  Detail & Related papers  (2025-03-01T07:02:30Z)
• Towards Unified Molecule-Enhanced Pathology Image Representation Learning via Integrating Spatial Transcriptomics [10.962389869127875]
  We introduce a Unified Molecule-enhanced Pathology Image REpresentationn Learning framework (UMPIRE)<n>UMPIRE aims to leverage complementary information from gene expression profiles to guide the multimodal pre-training.<n>We demonstrate that this molecular perspective provides a robust, task-agnostic training signal for learning pathology image embeddings.
  arXiv  Detail & Related papers  (2024-12-01T03:09:52Z)

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.