Knowledge-Primed Neural Networks

This repository provides a curated list of publications on Knowledge-Primed Neural Networks (KPNNs), with short summaries and links.

Note that in the literature also other names are used for the same concept, such as Visible Neural Networks (VNN), knowledge-guided neural networks, knowledge-based neural networks, ontology-based autoencoders, and biologically informed neural networks (BINNs). In the remainder of this repository, we use the term KPNN to collectively refer to all of these closely related approaches.

What are KPNNs?

KPNNs are a method in biological modelling and are artificial neural networks whose architecture is explicitly informed by a prior knowledge network, such as biological pathways or ontology networks. Network nodes correspond to nodes of the prior knowledge network, enabling a direct mapping between model components and domain concepts.

By combining this architecture with attribution methods, individual nodes can be ranked according to their contribution to the prediction task, which corresponds to their importance in the prior knowledge network. This makes model predictions interpretable, leading to scientific insights.

Literature Overview

Over the past years, a growing number of high-impact publications have explored Knowledge-Primed Neural Networks across multiple domains. Below is a curated list of representative papers in this research direction, organized by application field.

• Wang, 2026
  MULGONET: An interpretable neural network framework to integrate multi-omics data for cancer recurrence prediction and biomarker discovery. Fundamental Research.

• Han, 2025
  PathHDNN: a pathway hierarchical-informed deep neural network framework for predicting immunotherapy response and mechanism interpretation. Genome Medicine.

• Pickering, 2025
  Biology-informed neural networks learn nonlinear representations from omics data to improve genomic prediction and interpretability. arXiv.

• Joshi, 2025
  IRnet: Immunotherapy response prediction using pathway knowledge-informed graph neural network. Journal of Advanced Research.

• Huang, 2025
  Pathway-guided architectures for interpretable AI in biological research. Computational and Structural Biotechnology Journal.

• Nilson, 2025
  Biologically informed neural network models are robust to spurious interactions via self-pruning. bioRxiv.

• Vollmer, 2025
  Visible neural networks for multi-omics integration: a critical review. Frontiers in Artificial Intelligence.

• Vollmer, 2025
  Beyond the black box with biologically informed neural networks. Nature Reviews Genetics.

• Chen, 2024
  BioM2: biologically informed multi-stage machine learning for phenotype prediction using omics data. Briefings in Bioinformatics.

• McWeeney, 2024
  Graph Structured Neural Networks for Perturbation Biology. bioRxiv.

• Yang, 2024
  An interpretable survival model for diffuse large B-cell lymphoma patients using a biologically informed visible neural network. Computational and Structural Biotechnology Journal.

• Liu, 2024
  Prior knowledge-guided multilevel graph neural network for tumor risk prediction and interpretation via multi-omics data integration. Briefings in Bioinformatics.

• Ideker, 2024
  A deep learning model of tumor cell architecture elucidates response and resistance to CDK4/6 inhibitors. Nature Cancer.

• Li, 2024
  Using DeepSignalingFlow to mine signaling flows interpreting mechanism of synergy of cocktails. Systems Biology and Applications.

• Yachie, 2024
  DTox: A deep neural network-based in visio lens for large scale toxicogenomics data. The Journal of Toxicological Sciences.

• Chen, 2024
  DeepKEGG: a multi-omics data integration framework with biological insights for cancer recurrence prediction and biomarker discovery. Briefings in Bioinformatics.

• Heinemann, 2024
  DeepBINN: A tailored biologically-informed neural network for robust biomarker identification. 2024 11th IEEE Swiss Conference on Data Science (SDS).

• Wu, 2023
  PathExpSurv: pathway expansion for explainable survival analysis and disease gene discovery. BMC Bioinformatics.

• Ceccarelli, 2023
  MOViDA: multiomics visible drug activity prediction with a biologically informed neural network model. Bioinformatics.

• Cicek, 2023
  PiDeeL: metabolic pathway-informed deep learning model for survival analysis and pathological classification of gliomas. Bioinformatics.

• Lee, 2023
  PINNet: a deep neural network with pathway prior knowledge for Alzheimer's disease. Frontiers in Aging Neuroscience.

• Fortelny, 2023
  Reliable interpretability of biology-inspired deep neural networks. Systems Biology and Applications.

• Freitas, 2023
  A systematic review of biologically-informed deep learning models for cancer: fundamental trends for encoding and interpreting oncology data. BMC Bioinformatics.

• Chen, 2023
  CellTICS: an explainable neural network for cell-type identification and interpretation based on single-cell RNA-seq data. Briefings in Bioinformatics.

• Freitas, 2023
  A systematic review of biologically-informed deep learning models for cancer: fundamental trends for encoding and interpreting oncology data. BMC Bioinformatics.

• Loucera, 2023
  SigPrimedNet: A Signaling-Informed Neural Network for scRNA-seq Annotation of Known and Unknown Cell Types. Biology.

• Wang, 2023
  DeepGAMI: deep biologically guided auxiliary learning for multimodal integration and imputation to improve genotype-phenotype prediction. Genome Medicine.

• Lauffenburger, 2022
  Artificial neural networks enable genome-scale simulations of intracellular signaling. Nature Communications.

• Park, 2022
  DeepHisCoM: deep learning pathway analysis using hierarchical structural component models. Briefings in Bioinformatics.

• Hanczar, 2022
  GraphGONet: a self-explaining neural network encapsulating the Gene Ontology graph for phenotype prediction on gene expression. Bioinformatics.

• Mostafavi, 2022
  Obtaining genetics insights from deep learning via explainable artificial intelligence. Nature Reviews Genetics.

• He, 2022
  MPVNN: Mutated Pathway Visible Neural Network for interpretable prediction of cancer survival. Bioinformatics.

• Wu, 2022
  A Deep Neural Network for Gastric Cancer Prognosis Prediction Based on Biological Information Pathways. Journal of Oncology.

• Moore, 2022
  Knowledge-guided deep learning models of drug toxicity improve interpretation. Patterns.

• Wang, 2021
  ParsVNN: parsimony visible neural networks for uncovering cancer-specific and drug-sensitive genes and pathways. NAR Genomics and Bioinformatics.

• Roshchupkin, 2021
  GenNet framework: interpretable deep learning for predicting phenotypes from genetic data. Communications Biology.

• Hahn & Van Allen, 2021
  Biologically informed deep neural network for prostate cancer discovery. Nature.

• Lichtarge, 2021
  Using interpretable deep learning to model cancer dependencies. Bioinformatics.

• Zhao, 2021
  DeepOmix: A scalable and interpretable multi-omics deep learning framework and application in cancer survival analysis. Computational and Structural Biotechnology Journal.

• Ahn, 2021
  Classification and Functional Analysis between Cancer and Normal Tissues Using Explainable Pathway Deep Learning through RNA-Sequencing Gene Expression. International Journal of Molecular Sciences.

• Kohandel, 2021
  Systems biology informed neural networks predict response to PD-1 checkpoint blockade. Communications Biology.

• Hanczar, 2021
  Deep GONet: self-explainable deep neural network based on Gene Ontology for phenotype prediction from gene expression data. BMC Bioinformatics.

• Li, 2021
  Investigating the relevance of major signaling pathways in cancer survival using a biologically meaningful deep learning model. BMC Bioinformatics.

• Bock, 2020
  Knowledge-primed neural networks enable biologically interpretable deep learning on single-cell sequencing data. Genome Biology.

• Kang, 2020
  PAGE-Net: Interpretable and Integrative Deep Learning for Survival Analysis Using Histopathological Images and Genomic Data. Pacific Symposium on Biocomputing.

• Ideker, 2020
  Predicting Drug Response and Synergy Using a Deep Learning Model of Human Cancer Cells. Cancer Cell.

• Wang, 2021
  Varmole: a biologically drop-connect deep neural network model for prioritizing disease risk variants and genes. Bioinformatics.

• Liu, 2020
  Pathway-Guided Deep Neural Network toward Interpretable and Predictive Modeling of Drug Sensitivity. Journal of Chemical Information and Modeling.

• Gerstein, 2018
  Comprehensive functional genomic resource and integrative model for the human brain. Science.

• Ideker, 2018
  Using deep learning to model the hierarchical structure and function of a cell. Nature Methods.

• Kang, 2018
  PASNet: pathway-associated sparse deep neural network for prognosis prediction from high-throughput data. BMC Bioinformatics.

• Kang, 2018
  Cox-PASNet: pathway-based sparse deep neural network for survival analysis. IEEE BIBM.

Contributions and feedback

This list is intended to be a curated and evolving overview of publications on knowledge-primed, visible, and biologically informed neural network architectures.
If you believe that a publication should be included in this list, that an included publication does not fit the scope, or that any information has been recorded incorrectly, please open an issue and provide a short justification or reference.

Constructive feedback and corrections are very welcome!