Predicting multicellular function through multi-layer tissue networks

Abstract

Motivation: Understanding functions of proteins in specific human tissues is essential for insights into disease diagnostics and therapeutics, yet prediction of tissue-specific cellular function remains a critical challenge for biomedicine. Results: Here we present OhmNet, a hierarchy-aware unsupervised node feature learning approach for multi-layer networks. We build a multi-layer network, where each layer represents molecular interactions in a different human tissue. OhmNet then automatically learns a mapping of proteins, represented as nodes, to a neural embedding based low-dimensional space of features. OhmNet encourages sharing of similar features among proteins with similar network neighborhoods and among proteins activated in similar tissues. The algorithm generalizes prior work, which generally ignores relationships between tissues, by modeling tissue organization with a rich multiscale tissue hierarchy. We use OhmNet to study multicellular function in a multi-layer protein interaction network of 107 human tissues. In 48 tissues with known tissue-specific cellular functions, OhmNet provides more accurate predictions of cellular function than alternative approaches, and also generates more accurate hypotheses about tissue-specific protein actions. We show that taking into account the tissue hierarchy leads to improved predictive power. Remarkably, we also demonstrate that it is possible to leverage the tissue hierarchy in order to effectively transfer cellular functions to a functionally uncharacterized tissue. Overall, OhmNet moves from flat networks to multiscale models able to predict a range of phenotypes spanning cellular subsystems

In Proceedings of the 25th International Conference on Intelligent Systems for Molecular Biology (ISMB), 2017

References in corpus (5)

• Biological network comparison using graphlet degree distribution
• The physics of spreading processes in multilayer networks
• Random Walks on Multiplex Networks
• Centrality in Interconnected Multilayer Networks
• Data Fusion by Matrix Factorization

Cited by in corpus (107)

• A Comprehensive Survey on Graph Neural Networks
• Graph Neural Networks: A Review of Methods and Applications
• Machine Learning for Integrating Data in Biology and Medicine: Principles, Practice, and Opportunities
• Graph Self-Supervised Learning: A Survey
• Large-Scale Learnable Graph Convolutional Networks
• Self-Supervised Representation Learning: Introduction, Advances and Challenges
• Deep Graph Contrastive Representation Learning
• Graph Embedding on Biomedical Networks: Methods, Applications, and Evaluations
• Everything is Connected: Graph Neural Networks
• Revisiting Graph Neural Networks: All We Have is Low-Pass Filters
• PyTorch-BigGraph: A Large-scale Graph Embedding System
• DeepGCNs: Making GCNs Go as Deep as CNNs
• Multilayer Networks in a Nutshell
• SIGN: Scalable Inception Graph Neural Networks
• Design Space for Graph Neural Networks
• Network Enhancement: a general method to denoise weighted biological networks
• Graph Barlow Twins: A self-supervised representation learning framework for graphs
• How to Find Your Friendly Neighborhood: Graph Attention Design with Self-Supervision
• Current and future directions in network biology
• DeepGCNs: Can GCNs Go as Deep as CNNs?
• Measuring and Relieving the Over-smoothing Problem for Graph Neural Networks from the Topological View
• The Evolution of Distributed Systems for Graph Neural Networks and their Origin in Graph Processing and Deep Learning: A Survey
• Automated Machine Learning on Graphs: A Survey
• Anomaly Detection in Dynamic Graphs: A Comprehensive Survey
• Auto-GNN: Neural Architecture Search of Graph Neural Networks
• Janossy Pooling: Learning Deep Permutation-Invariant Functions for Variable-Size Inputs
• On the Trade-off between Over-smoothing and Over-squashing in Deep Graph Neural Networks
• MultiVERSE: a multiplex and multiplex-heterogeneous network embedding approach
• GeniePath: Graph Neural Networks with Adaptive Receptive Paths
• GNN-FiLM: Graph Neural Networks with Feature-wise Linear Modulation
• Unsupervised Graph Poisoning Attack via Contrastive Loss Back-propagation
• Graph Normalizing Flows
• Dirichlet Energy Constrained Learning for Deep Graph Neural Networks
• MGCN: Semi-supervised Classification in Multi-layer Graphs with Graph Convolutional Networks
• Tensor Graph Convolutional Networks for Multi-relational and Robust Learning
• AGL: a Scalable System for Industrial-purpose Graph Machine Learning
• Nine Quick Tips for Analyzing Network Data
• Boosting-GNN: Boosting Algorithm for Graph Networks on Imbalanced Node Classification
• Meta-Graph: Few Shot Link Prediction via Meta Learning
• Propositionalization and Embeddings: Two Sides of the Same Coin
• Multilayer Network Modeling of Integrated Biological Systems
• Understanding the Message Passing in Graph Neural Networks via Power Iteration Clustering
• GraphCL: Contrastive Self-Supervised Learning of Graph Representations
• Pre-Training Graph Neural Networks for Generic Structural Feature Extraction
• Generative Causal Explanations for Graph Neural Networks
• Relational Self-Supervised Learning on Graphs
• Probabilistic Dual Network Architecture Search on Graphs
• Distilling Knowledge from Graph Convolutional Networks
• DGNN: Decoupled Graph Neural Networks with Structural Consistency between Attribute and Graph Embedding Representations
• Reinforced Neighborhood Selection Guided Multi-Relational Graph Neural Networks
• Residual or Gate? Towards Deeper Graph Neural Networks for Inductive Graph Representation Learning
• Learned Low Precision Graph Neural Networks
• Advancing GraphSAGE with A Data-Driven Node Sampling
• Graph Similarity Description: How Are These Graphs Similar?
• A Simple Proof of the Universality of Invariant/Equivariant Graph Neural Networks
• FairEGM: Fair Link Prediction and Recommendation via Emulated Graph Modification
• SAC: Accelerating and Structuring Self-Attention via Sparse Adaptive Connection
• MultiNet: A Scalable Multilayer Network Embedding Framework
• Sampling methods for efficient training of graph convolutional networks: A survey
• Understanding Graph Neural Networks from Graph Signal Denoising Perspectives
• SGAS: Sequential Greedy Architecture Search
• Graph Meta Learning via Local Subgraphs
• Resurrecting Label Propagation for Graphs with Heterophily and Label Noise
• General Partial Label Learning via Dual Bipartite Graph Autoencoder
• Edge Dithering for Robust Adaptive Graph Convolutional Networks
• Resilience in multiplex networks by addition of cross-repulsive links
• Revisiting Graph Convolutional Network on Semi-Supervised Node Classification from an Optimization Perspective
• Identifying Illicit Accounts in Large Scale E-payment Networks -- A Graph Representation Learning Approach
• Zoo Guide to Network Embedding
• CARL-G: Clustering-Accelerated Representation Learning on Graphs
• Graphlets versus node2vec and struc2vec in the task of network alignment
• Hierarchical Message-Passing Graph Neural Networks
• Continual Graph Learning: A Survey
• Link Prediction using Graph Neural Networks for Master Data Management
• Feature Propagation on Graph: A New Perspective to Graph Representation Learning
• Deep Node Ranking for Neuro-symbolic Structural Node Embedding and Classification
• SkipGNN: Predicting Molecular Interactions with Skip-Graph Networks
• Deep learning in bioinformatics: introduction, application, and perspective in big data era
• Self-Supervised Graph Representation Learning via Global Context Prediction
• Adaptive Multi-layer Contrastive Graph Neural Networks
• $p$-Laplacian Based Graph Neural Networks
• On Self-Distilling Graph Neural Network
• Representation Learning of Graphs Using Graph Convolutional Multilayer Networks Based on Motifs
• Hierarchical Protein Function Prediction with Tail-GNNs
• GraphScale: A Framework to Enable Machine Learning over Billion-node Graphs
• HighwayGraph: Modelling Long-distance Node Relations for Improving General Graph Neural Network
• Graph Representation Learning on Tissue-Specific Multi-Omics
• GNNSampler: Bridging the Gap between Sampling Algorithms of GNN and Hardware
• The Split Matters: Flat Minima Methods for Improving the Performance of GNNs
• Computational strategies for cross-species knowledge transfer
• EWasteNet: A Two-Stream Data Efficient Image Transformer Approach for E-Waste Classification
• Deep Contextual Learners for Protein Networks
• Graph Pruning for Model Compression
• Probing Negative Sampling Strategies to Learn GraphRepresentations via Unsupervised Contrastive Learning
• GAIN: Graph Attention & Interaction Network for Inductive Semi-Supervised Learning over Large-scale Graphs
• Graphs, Entities, and Step Mixture
• GFCN: A New Graph Convolutional Network Based on Parallel Flows
• Scalable Explanation of Inferences on Large Graphs
• Lovasz Convolutional Networks
• An Uncoupled Training Architecture for Large Graph Learning
• DuoGNN: Topology-aware Graph Neural Network with Homophily and Heterophily Interaction-Decoupling
• Dynamic Embedding on Textual Networks via a Gaussian Process
• LSP : Acceleration and Regularization of Graph Neural Networks via Locality Sensitive Pruning of Graphs
• Isometric Graph Neural Networks
• Network Representation Learning: From Traditional Feature Learning to Deep Learning
• PiNet: Attention Pooling for Graph Classification
• Dual GNNs: Graph Neural Network Learning with Limited Supervision