SCR2-ST: Combine Single Cell with Spatial Transcriptomics for Efficient Active Sampling via Reinforcement Learning

2025-12-15

AI Summary

Spatial transcriptomics (ST) data acquisition is costly, and fixed-grid sampling introduces redundancy in morphologically or biologically homogeneous regions, exacerbating data scarcity. To address this, we propose the first single-cell-guided reinforcement learning framework for active ST sampling. Our method leverages embeddings from a single-cell foundation model to encode prior biological knowledge, integrates spatial density modeling with multi-source data alignment, and enables intelligent, budget-constrained spot selection. Furthermore, we design a hybrid regression-retrieval prediction network coupled with cell-type-aware soft-label supervision to enhance gene expression reconstruction accuracy under extreme data scarcity (<10% sampled spots). Evaluated on three public ST datasets, our approach significantly outperforms existing methods in both sampling efficiency and reconstruction fidelity, achieving state-of-the-art performance across all benchmarks.

Technology Category

• Machine Learning: Active Learning
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Spatial transcriptomics (ST) is an emerging technology that enables researchers to investigate the molecular relationships underlying tissue morphology. However, acquiring ST data remains prohibitively expensive, and traditional fixed-grid sampling strategies lead to redundant measurements of morphologically similar or biologically uninformative regions, thus resulting in scarce data that constrain current methods. The well-established single-cell sequencing field, however, could provide rich biological data as an effective auxiliary source to mitigate this limitation. To bridge these gaps, we introduce SCR2-ST, a unified framework that leverages single-cell prior knowledge to guide efficient data acquisition and accurate expression prediction. SCR2-ST integrates a single-cell guided reinforcement learning-based (SCRL) active sampling and a hybrid regression-retrieval prediction network SCR2Net. SCRL combines single-cell foundation model embeddings with spatial density information to construct biologically grounded reward signals, enabling selective acquisition of informative tissue regions under constrained sequencing budgets. SCR2Net then leverages the actively sampled data through a hybrid architecture combining regression-based modeling with retrieval-augmented inference, where a majority cell-type filtering mechanism suppresses noisy matches and retrieved expression profiles serve as soft labels for auxiliary supervision. We evaluated SCR2-ST on three public ST datasets, demonstrating SOTA performance in both sampling efficiency and prediction accuracy, particularly under low-budget scenarios. Code is publicly available at: https://github.com/hrlblab/SCR2ST

Problem

• Integrates single-cell data to guide spatial transcriptomics sampling efficiently
• Uses reinforcement learning for active sampling under constrained sequencing budgets
• Improves prediction accuracy with hybrid regression-retrieval network for sparse data

Innovation

• Reinforcement learning for active sampling
• Hybrid regression-retrieval prediction network
• Single-cell guided biologically grounded rewards