PCGS: biomarker and risk group identification for Pediatric Cancers via explainable Graph neural networks with Shapley values
- 1Indiana University School of Medicine
Abstract
Improvements in data availability, sharing, and integration, together with the development of explainable artificial intelligence (XAI) techniques, are advancing precision medicine for pediatric cancer by facilitating diagnosis, biomarker discovery, and drug development. Data sharing commons and initiatives like the Childhood Cancer Data Initiative (CCDI) provide access to pediatric-specific genomic and clinical data cohorts and improve data availability for pediatric cancer research. Based on CCDI, a scalable AI platform, Graph Artificial Intelligence for Pediatric Oncology (GAIPO), integrates various data modalities from bulk and single-cell omics data to clinical information. Such multi-modal data facilitates the training and development of advanced XAI models for pediatric cancers. We then developed an end-to-end multi-modality framework, PCGS, for pediatric cancer by incorporating omics-specific representation learning via GNN models with cross-attention fusion and multi-objective learning for downstream tasks such as classification, clustering, and survival analysis. This framework outperforms previous supervised multi-omics integration baseline approaches based on glioma and Wilms tumor cohorts and enables GNN model explainability via Shapley value-based feature attribution approaches to explain the contributions of gene-level features across various biomedical tasks, including classification and survival. Given specific background samples (e.g., age groups, sex, grades) as baselines, this explainable GNN model estimates and ranks the importance scores for input features from each omics modality. It identifies background-specific key features for biomarker discovery, risk group identification, and survival analysis in glioma and Wilms tumor, with potential applicability to other pediatric cancers.
Related
- Explainable computational framework for subtype classification and biomarker identification in head and neck squamous cell carcinoma using machine learning
- METTL1 gene dosage and catalytic activity define a targetable vulnerability in pediatric cancers
- WRN helicase addiction in MSI-H colorectal cancer: biomarker-guided pharmacology, resistance evolution, and implications for other gastrointestinal cancers
- The application of explainable machine learning for the identification of MASLD in non-obese populations
- Multi-omics profiling uncovers diagnostic biomarker panel, progression-associated subtypes, and prognostic signature in pediatric mitochondrial disease
BibTeX
@misc{shi2026pcgs,
title = {{PCGS: biomarker and risk group identification for Pediatric Cancers via explainable Graph neural networks with Shapley values}},
author = {Z. Shi and A. Budhkar and W. Amin and K. E. Pollok and J. Su and K. Huang},
howpublished = {medRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.27.26361540},
url = {https://www.medrxiv.org/content/10.64898/2026.08.27.26361540v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - PCGS: biomarker and risk group identification for Pediatric Cancers via explainable Graph neural networks with Shapley values
AU - Z. Shi
AU - A. Budhkar
AU - W. Amin
AU - K. E. Pollok
AU - J. Su
AU - K. Huang
PY - 2026
DA - 2026/08/27
PB - medRxiv
DO - 10.64898/2026.08.27.26361540
UR - https://www.medrxiv.org/content/10.64898/2026.08.27.26361540v1
AB - Improvements in data availability, sharing, and integration, together with the development of explainable artificial intelligence (XAI) techniques, are advancing precision medicine for pediatric cancer by facilitating diagnosis, biomarker discovery, and drug development. Data sharing commons and initiatives like the Childhood Cancer Data Initiative (CCDI) provide access to pediatric-specific genomic and clinical data cohorts and improve data availability for pediatric cancer research. Based on CCDI, a scalable AI platform, Graph Artificial Intelligence for Pediatric Oncology (GAIPO), integrates various data modalities from bulk and single-cell omics data to clinical information. Such multi-modal data facilitates the training and development of advanced XAI models for pediatric cancers. We then developed an end-to-end multi-modality framework, PCGS, for pediatric cancer by incorporating omics-specific representation learning via GNN models with cross-attention fusion and multi-objective learning for downstream tasks such as classification, clustering, and survival analysis. This framework outperforms previous supervised multi-omics integration baseline approaches based on glioma and Wilms tumor cohorts and enables GNN model explainability via Shapley value-based feature attribution approaches to explain the contributions of gene-level features across various biomedical tasks, including classification and survival. Given specific background samples (e.g., age groups, sex, grades) as baselines, this explainable GNN model estimates and ranks the importance scores for input features from each omics modality. It identifies background-specific key features for biomarker discovery, risk group identification, and survival analysis in glioma and Wilms tumor, with potential applicability to other pediatric cancers.
ER -