Article

Class-Support Mismatch Dominates Protocol-Driven Optimism in Spatial Transcriptomics, with a Larger Residual Penalty for Graph Neural Networks

2026-08-03

Abstract excerpt

<title>Abstract</title> <p>Motivation: Spatial transcriptomics (ST) benchmarks are routinely reported to be “inflated” by randomly interleaved cross-validation, with the inflation attributed to spatial leakage. That attribution has never been tested. The random-versus-block performance gap simultaneously changes the training-set size, the set of classes available in training and test, the amount of message passin...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: d56c14d0-27fc-5f74-a862-8d14f728c587
DOI: 10.21203/rs.3.rs-10544415/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Class-Support Mismatch Dominates Protocol-Driven Optimism in Spatial Transcriptomics, with a Larger Residual Penalty for Graph Neural NetworksDOI 10.21203/rs.3.rs-10544415/v1Read

Pillbox: A Leakage-Aware Foundation-Model Predictor and Lineage-Ceiling Diagnostic for Cancer Drug ResponseSemantic proximity Read

Comprehensive benchmarking of batch integration methods for spatial transcriptomics using a large-scale cancer atlasSemantic proximity Read

A method-independent recoverability ceiling for transcript-to-cell assignment in imaging-based spatial transcriptomicsSemantic proximity Read

Towards a Better Understanding of Batch Effects in Spatial Transcriptomics: Definition and Method EvaluationSemantic proximity Read

Three Classes of Confound in Gene-Regulatory-Network Inference: A Systematic Audit and Open-Source Diagnostic ToolkitSemantic proximity Read

Experimental Design Dominates Model Architecture in Multi-Model Cancer Classification: A Translational Bioinformatics Analysis Across Small-n and Large-n RegimesSemantic proximity Read

Cell-CLIP: Multi-algorithm causal discovery of directed cell program interaction networks from single-cell transcriptomicsSemantic proximity Read

GraphTransformer-CoGNN: A two-stage dynamic graph learning framework for label-scarce cell-type annotation in spatial transcriptomicsSemantic proximity Read

LNGCN: a distance-aware continuous-time graph framework for PPI prioritization in experimental screeningSemantic proximity Read

Non-gradient spatial organisation at tumour–immune interfaces revealed by operator-based analysisSemantic proximity Read

StickForStats: automated statistical assumption validation for reproducible computational biologySemantic proximity Read

When Simpler Models Win: A Calibration Benchmark of scGPT for Cell-Type AnnotationSemantic proximity Read