Zero-Shot Metabolite Prediction from Gene Expression via Physics-Informed Graph Neural Networks

bioRxiv (Cold Spring Harbor Laboratory)

Abstract

ABSTRACT Predicting metabolite concentrations from gene expression is instrumental for linking regulatory programs to metabolic phenotypes. Prior approaches rely on static enzyme–metabolite mappings and often omit data-driven learning or biochemical constraints, limiting their ability to generalize to new metabolites. We present GAZE (Graph Attention for Zero-shot metabolite Estimation), a physics-informed graph neural network that integrates enzyme expression, Enzyme Commission functional embeddings, and ChemBERTa metabolite descriptors within a unified metabolic graph (5,414 nodes, 16,307 edges). A Metabolite-Conditioned Reader uses each metabolite’s SMILES embedding to query learned pathway representations, enabling zero-shot prediction with no metabolite-specific parameters. We evaluate GAZE in three scenarios: (i) standard cross-validation on the Cancer Atlas of Metabolic Profiles (18,044 genes, 180 metabolites, 867 cell lines), achieving R 2 = 0.816; (ii) leave-one-metabolite-out (LOMO) zero-shot evaluation across 50 held-out metabolites, where the physics-informed variant halves the median R 2 deficit relative to a standard GNN baseline (−0.34 vs. −0.72), with 30% of unseen metabolites achieving positive R 2 ; and (iii) external validation on an independent clear cell renal cell carcinoma tissue cohort (220 samples), where GAZE achieves median Spearman ρ = 0.330 across 214 metabolites without fine-tuning. GAZE outperforms scCellFie, MEBOCOST, and UnitedMet across all evaluation settings.

Authors 3

1. Claudio Novella-Rausell Aachen Department of Human Genetics Department of Internal Medicine (Nephrology) & Einthoven Laboratory of Vascular and Regenerative Medicine Novo Nordisk Foundation Center for Stem Cell Medicine (reNEW)

   Leiden University · Leiden University Medical Center · Novo Nordisk Foundation Center for Stem Cell Medicine

   Affiliation as printed

   Department of Human Genetics, Leiden University Medical Centre, 2333 ZA Leiden, the Netherlands

   Department of Internal Medicine (Nephrology) & Einthoven Laboratory of Vascular and Regenerative Medicine, Leiden University Medical Center, Leiden, The Netherlands

   The Novo Nordisk Foundation Center for Stem Cell Medicine (reNEW), Leiden University Medical Center, Leiden, The Netherlands

2. Ton J. Rabelink Aachen Department of Internal Medicine (Nephrology) & Einthoven Laboratory of Vascular and Regenerative Medicine Novo Nordisk Foundation Center for Stem Cell Medicine (reNEW)

   Leiden University · Leiden University Medical Center · Novo Nordisk Foundation Center for Stem Cell Medicine

   Affiliation as printed

   Department of Internal Medicine (Nephrology) & Einthoven Laboratory of Vascular and Regenerative Medicine, Leiden University Medical Center, Leiden, The Netherlands

   The Novo Nordisk Foundation Center for Stem Cell Medicine (reNEW), Leiden University Medical Center, Leiden, The Netherlands

3. Ahmed Essam Mahfouz corresponding Aachen Department of Human Genetics Leiden Computational Biology Center

   Leiden University · Leiden University Medical Center · Delft University of Technology

   Affiliation as printed

   Delft Bioinformatics Lab, Delft University of Technology, Delft, the Netherlands

   Department of Human Genetics, Leiden University Medical Centre, 2333 ZA Leiden, the Netherlands

   Leiden Computational Biology Center, Leiden University Medical Center, Leiden, the Netherlands