Branch-Wise Regularization for Heterogeneous GNN-Based MALDI-TOF AMR Prediction with Biomarker-Consensus Edges
- 1National Chung Hsing University
Abstract
This paper predicts antimicrobial resistance (AMR) from matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectra using DRIAMS, evaluated across 13 species-antibiotic datasets spanning sensitive/resistant imbalance ratios of 2.95:1 to 141.21:1 and raw spectral similarities of 0.9166-0.9946. We propose HetSAGE, a heterogeneous graph neural network (GNN) that combines full-spectrum "raw" edges with consensus "biomarker" edges derived from a 4-method feature-selection vote and applies per-edge-type dropout to regularize the two views separately. HetSAGE outperforms a plain multilayer perceptron (MLP) only when spectral similarity is high, consistent with oversmoothing effects reported in the broader GNN literature; the biomarker view converges on features close to independently established clinical markers: within {+/-}1 Da in Staphylococcus aureus, Oxacillin, and within range in Klebsiella pneumoniae, Meropenem, confirming it is not a black box. At the same time, per-edge-type dropout protects this smaller signal from dilution by regularization tuned for the noisier raw spectrum. Together, these results indicate that heterogeneous graph structure is not a universal win for MALDI-TOF AMR prediction but a conditional one, tied to a measurable property of the data, spectral similarity, rather than to architecture alone. This caveat is largely unaddressed in current DRIAMS-based GNN work. It provides a concrete signal for practitioners deciding when a graph model is worth the added complexity over a plain MLP.
Related
- MarkerScout: A Disease-Agnostic Machine Learning Framework for Biomarker Prediction from Multi-Scale Mechanistic Models
- Heterogeneous Graph Contrastive Learning for Drug-Gene-Disease Motif Prediction
- Mechanical characterization of glioblastoma cell – substrate interaction by fading regularization method
- Reflections on the ESC consensus statement on inflammation in coronary syndromes: From mechanistic certainty to practical cardiology
- The Management of Type 1 Diabetes in Adults. The Updated 2026 Consensus Report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD)
BibTeX
@misc{tsai2026branchwise,
title = {{Branch-Wise Regularization for Heterogeneous GNN-Based MALDI-TOF AMR Prediction with Biomarker-Consensus Edges}},
author = {W.-Y. Tsai and Y.- T. Shih and Y.- H. Chen},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.24.754085},
url = {https://www.biorxiv.org/content/10.64898/2026.09.24.754085v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Branch-Wise Regularization for Heterogeneous GNN-Based MALDI-TOF AMR Prediction with Biomarker-Consensus Edges
AU - W.-Y. Tsai
AU - Y.- T. Shih
AU - Y.- H. Chen
PY - 2026
DA - 2026/09/24
PB - bioRxiv
DO - 10.64898/2026.09.24.754085
UR - https://www.biorxiv.org/content/10.64898/2026.09.24.754085v1
AB - This paper predicts antimicrobial resistance (AMR) from matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectra using DRIAMS, evaluated across 13 species-antibiotic datasets spanning sensitive/resistant imbalance ratios of 2.95:1 to 141.21:1 and raw spectral similarities of 0.9166-0.9946. We propose HetSAGE, a heterogeneous graph neural network (GNN) that combines full-spectrum "raw" edges with consensus "biomarker" edges derived from a 4-method feature-selection vote and applies per-edge-type dropout to regularize the two views separately. HetSAGE outperforms a plain multilayer perceptron (MLP) only when spectral similarity is high, consistent with oversmoothing effects reported in the broader GNN literature; the biomarker view converges on features close to independently established clinical markers: within {+/-}1 Da in Staphylococcus aureus, Oxacillin, and within range in Klebsiella pneumoniae, Meropenem, confirming it is not a black box. At the same time, per-edge-type dropout protects this smaller signal from dilution by regularization tuned for the noisier raw spectrum. Together, these results indicate that heterogeneous graph structure is not a universal win for MALDI-TOF AMR prediction but a conditional one, tied to a measurable property of the data, spectral similarity, rather than to architecture alone. This caveat is largely unaddressed in current DRIAMS-based GNN work. It provides a concrete signal for practitioners deciding when a graph model is worth the added complexity over a plain MLP.
ER -