HESCAPE PyArrow: A Multimodal Benchmark for Spatial Transcriptomics

HESCAPE is a large-scale benchmark for multimodal learning in spatial transcriptomics, hosted by Peng-AI. The dataset is organized into multiple panel-specific configurations, including human-5k-panel, human-breast-panel, and human-colon-panel. It was last updated on September 16, 2025.

Use Cases

• Benchmarking multimodal learning models based on combined histology and spatial transcriptomics data.
• Training contrastive pretraining models based on the H&E and spatial data pairing.
• Developing panel-specific analysis tools based on the available human tissue configurations.
• Evaluating model performance on distinct tissue types based on the separate breast, colon, and lung panel datasets.

Strengths

• Dataset is organized into multiple panel-specific configurations, suggesting structured data for different tissue types.
• The benchmark is described as large-scale, implying a substantial volume of data.
• Last updated on September 16, 2025, indicating recent maintenance.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Peng-AI

Freshness

Last updated 2025-09-16 18:00:55.