Sa Li

I am a Ph.D. candidate at Oakland University working on graph neural networks for cancer driver gene identification. My research focuses on topology-adaptive GNNs, multi-omics data integration, and interpretable deep learning for biologically grounded discovery.

Research Interests

Graph Neural Networks · Computational Cancer Genomics · Multi-Omics Integration · Interpretable AI · Network Biology

Research Project & Funding

This research is supported by the U.S. National Science Foundation (NSF) under Grant No. 2245805. The project focuses on developing topology-adaptive and interpretable graph neural networks for cancer driver gene identification through large-scale multi-omics data integration and biological network modeling.

The proposed research advances foundational methods in graph representation learning, computational cancer genomics, and explainable artificial intelligence, with the goal of enabling biologically grounded discovery and improving the interpretability of deep learning models in precision medicine.

Teaching Experience

Instructor
Southwestern Michigan College, Dowagiac, MI  |  Fall 2021
ISYS–115: Introduction to Programming Logic and Design

Teaching Assistant
Oakland University
Courses assisted: Programming Languages, Computer Networks, Computer Systems, Software Security, Object-Oriented Computing, and Software Modeling.

Publications

• Sa, L., Shader, J., Bhattacharya, A., Ma, T.
  Learning Pretrained Graph Representations for Pathway Network Inference.
  Proceedings of the 19th International Conference on Bioinformatics Models, Methods and Algorithms (BIOINFORMATICS 2026),
  Oral Presentation, 2026.
• Sa, L., Shader, J., Ma, T.
  PERGAT: Pretrained Embeddings of Graph Neural Networks for miRNA–Cancer Association Prediction.
  Proceedings of the IEEE International Conference on Bioinformatics and Biomedicine (BIBM 2024), Lisbon, Portugal, 2024.
  DOI: 10.1109/BIBM62325.2024.10822135
• Sa, L., Shader, J., Bhattacharya, A., Ma, T.
  Graph Neural Network with Pretrained Embeddings for Cancer and miRNA Regulator Discovery.
  Proceedings of the 16th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM BCB 2025).
  Article No. 78, Published December 10, 2025.
  DOI: 10.1145/3768322.3769029
• Sa, L., Shader, J., Bhattacharya, A., Ma, T.
  Integration of Multi-Omics Data with Topology Adaptive Graph Convolutional Network for Cancer Driver Gene Identification.
  IEEE Transactions on Computational Biology and Bioinformatics,
  Early Access, Nov. 27, 2025.
  DOI: 10.1109/TCBBIO.2025.3636976
  PMID: 41308109

Manuscripts Under Review

• Sa, L., Ma, T.
  Learning Interpretable Pathway Representations with Spectral Graph Neural Networks.
  Submitted to Bioinformatics, under review.
• Sa, L., Ma, T.
  Spectral Graph Embedding Framework for Multi-Omics Cancer Driver Gene Prioritization.
  Submitted to Bioinformatics, under review.
• Sa, L., Ma, T.
  Pathway Representation Learning for Interpretable Graph Neural Networks.
  Submitted to IEEE Access, under review.
• Sa, L., Ma, T.
  Multi-Omics Integration with Spectral GNNs for Identifying Cancer Driver Genes.
  Submitted to IEEE Transactions on Computational Biology and Bioinformatics, under review.

Hobbies

AlphaGo · Basketball · Skateboarding · Acoustic Guitar