Research map / Probabilistic and scientific modeling / Molecules and proteins
HEIST: A Graph Foundation Model for Spatial Transcriptomics and Proteomics Data
Hiren Madhu, João Felipe Rocha, Tinglin Huang, Siddharth Viswanath, Smita Krishnaswamy, Rex Ying
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| fine-tune the HEIST decoder instead of zero-shot use | HEIST · gene imputation, Skin | Pearson correlation: 0.350 -> 0.807 | Improved |
| add hierarchical modeling with cross-level message passing (full HEIST vs removing cross message passing) (ablation) | HEIST · Charville outcome | AUC-ROC: 0.625 -> 0.861 | Improved |
| add pre-training (ablation) | HEIST · Charville outcome | AUC-ROC: 0.500 -> 0.861 | Improved |
| add spatial cell graph level (hierarchy) (ablation) | HEIST · SEA-AD cell classification | F1: 0.179 -> 0.995 | Improved |
| add gene co-expression level (hierarchy) (ablation) | HEIST · Skin imputation | Pearson correlation: 0.173 -> 0.807 | Improved |
| add masked autoencoder objective (ablation) | HEIST · SEA-AD cell classification | F1: 0.162 -> 0.995 | Improved |
| add contrastive objective (ablation) | HEIST · SEA-AD cell classification | F1: 0.966 -> 0.995 | Improved |
| add positional encodings (ablation) | HEIST · SEA-AD cell classification | F1: 0.220 -> 0.995 | Improved |
| add orthogonal regularization (ablation) | HEIST · Skin imputation | Pearson correlation: 0.646 -> 0.807 | Improved |
| use sinusoidal positional encodings instead of Laplacian or random-walk PEs (ablation) | HEIST · ablation of positional encoding types | downstream performance | Improved |
| use adaptive instead of constant co-expression threshold tau (ablation) | HEIST · ablation of graph threshold | downstream performance | Improved |
| pass raw features through MLP before adding (ablation) | HEIST · gene imputation | imputation performance | Hurt |
Closest papers by meaning
- HEST-1k: A Dataset For Spatial Transcriptomics and Histology Image Analysis (NeurIPS 2024 · similarity 0.89)
- Diffusion Generative Modeling for Spatially Resolved Gene Expression Inference from Histology Images (ICLR 2025 · similarity 0.88)
- Scalable Generation of Spatial Transcriptomics from Histology Images via Whole-Slide Flow Matching (ICML 2025 · similarity 0.87)
- Towards Scalable Context-Aware Single-Cell Spatial Transcriptomics Prediction from Histology Images (NeurIPS 2026 · similarity 0.87)
- Preserving DEG Rankings for Gene Discovery in Histology-Based Spatial Gene Expression Prediction (NeurIPS 2026 · similarity 0.86)
- MERGE: Multi-faceted Hierarchical Graph-based GNN for Gene Expression Prediction from Whole Slide Histopathology Images (CVPR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
