Abstract:High-throughput technologies are becoming increasingly important in discovering prognostic biomarkers and in identifying novel drug targets. With Mammaprint, Oncotype DX, and many other prognostic molecular signatures breast cancer is one of the paradigmatic examples of the utility of high-throughput data to deliver prognostic biomarkers, that can be represented in a form of a rather short gene list. Such gene lists can be obtained as a set of features (genes) that are important for the decisions of a Machine Learning (ML) method applied to high-dimensional gene expression data. Several studies have identified predictive gene lists for patient prognosis in breast cancer, but these lists are unstable and have only a few genes in common. Instability of feature selection impedes biological interpretability: genes that are relevant for cancer pathology should be members of any predictive gene list obtained for the same clinical type of patients. Stability and interpretability of selected features can be improved by including information on molecular networks in ML methods. Graph Convolutional Neural Network (GCNN) is a contemporary deep learning approach applicable to gene expression data structured by a prior knowledge molecular network. Layer-wise Relevance Propagation (LRP) and SHapley Additive exPlanations (SHAP) are methods to explain individual decisions of deep learning models. We used both GCNN+LRP and GCNN+SHAP techniques to construct feature sets by aggregating individual explanations. We suggest a methodology to systematically and quantitatively analyze the stability, the impact on the classification performance, and the interpretability of the selected feature sets. We used this methodology to compare GCNN+LRP to GCNN+SHAP and to more classical ML-based feature selection approaches. Utilizing a large breast cancer gene expression dataset we show that, while feature selection with SHAP is useful in applications where selected features have to be impactful for classification performance, among all studied methods GCNN+LRP delivers the most stable (reproducible) and interpretable gene lists.

Gene selection of microarray data using heatmap analysis and Graph Neural Network

Feature selection of microarray data using multidimensional graph neural network and supernode hierarchical clustering

GAMB-GNN: Graph Neural Networks learning from gene structure relations and Markov Blanket ranking for cancer classification in microarray data

Parameters Selection in Gene Selection Using Gaussian Kernel Support Vector Machines by Genetic Algorithm

Hierarchical System of Gene Selection Based on Deep Learning and Ensemble Approach

Classification of human cancer diseases by gene expression profiles

Gene selection and sample classification on microarray data based on adaptive genetic algorithm/k-nearest neighbor method

Gene Selection for Cancer Classification using Support Vector Machines

An efficient feature selection and classification system for microarray cancer data using genetic algorithm and deep belief networks

Gene selection for cancer identification: a decision tree model empowered by particle swarm optimization algorithm

Deep Learning Based Tumor Type Classification Using Gene Expression Data

A hybrid feature selection approach for Microarray datasets using graph theoretic-based method

A novel parallel feature rank aggregation algorithm for gene selection applied to microarray data classification

Comparisons of Graph Neural Networks on Cancer Classification Leveraging a Joint of Phenotypic and Genetic Features

Gene selection and classification for cancer microarray data based on machine learning and similarity measures

Signature Genes Selection and Functional Analysis of Phenotypes: A Comparative Study

Gene selection for cancer classification using a hybrid of univariate and multivariate feature selection methods

Identification and segregation of genes with improved recurrent neural network trained with optimal gene level and mutation level features

Analysis of Gene Regulatory Networks from Gene Expression Using Graph Neural Networks

Hybrid gene selection approach using XGBoost and multi-objective genetic algorithm for cancer classification

Stable feature selection utilizing Graph Convolutional Neural Network and Layer-wise Relevance Propagation for biomarker discovery in breast cancer