Multi-Omic Graph Diagnosis (MOGDx) : A data integration tool to perform classification tasks for heterogeneous diseases

Abstract

Heterogeneity in human diseases presents challenges in diagnosis and treatments due to the broad range of manifestations and symptoms. With the rapid development of labelled multi-omic data, integrative machine learning methods have achieved breakthroughs in treatments by redefining these diseases at a more granular level. These approaches often have limitations in scalability, oversimplification, and handling of missing data. In this study, we introduce Multi-Omic Graph Diagnosis (MOGDx), a flexible command line tool for the integration of multi-omic data to perform classification tasks for heterogeneous diseases. MOGDx is a network integrative method that combines patient similarity networks with a reduced vector representation of genomic data. The reduced vector is derived from the latent embeddings of an auto-encoder and the combined network is fed into a graph convolutional network for classification. MOGDx was evaluated on three datasets from the cancer genome atlas for breast invasive carcinoma, kidney cancer, and low grade glioma. MOGDx demonstrated state-of-the-art performance and an ability to identify relevant multi-omic markers in each task. It did so while integrating more genomic measures with greater patient coverage compared to other network integrative methods. MOGDx is available to download from https://github.com/biomedicalinformaticsgroup/MOGDx. Overall, MOGDx is a promising tool for integrating multi-omic data, classifying heterogeneous diseases, and interpreting genomic markers.

An Integrative Network Approach for Longitudinal Stratification in Parkinson’s Disease

“…The GCN-MME is trained under the semi-supervised setting for GNN outlined by Hamilton (2020) [7]. For a detailed description of the MOGDx architecture, please refer to Ryan et al 2023 [22].…”

“…MOGDx achieves this by utilising SNF and imputation to retain patient nodes. Moreover, including patients missing in one or more modalities does not result in a large degradation in performance [22]. MOGDx provides a high level of interpretability.…”

“…For example, DNAm was normalised using the wateRmelon package in R [20]. See Ryan et al (2023) for more detail on how specific genomic modalities were handled [22]. In general, processing included the removal of uninformative or missing features, normalisation, imputation of missing values and conversion from categorical features to numerical features.…”

“…This approach uses a network taxonomy to combine patient similarity matrices into a single network and perform node classification using a Graph Convolutional Network (GCN). The performance of MOGDx was benchmarked on cancer data and achieved state-of-the-art performance compared to similar research (Ryan et al, 2023).…”

“…The node feature matrix and fused PSN are combined and input into the GCN for classification. For a more detailed description of the MOGDx architecture, please refer to Ryan et al (2023) MOGDx is a preferred tool to perform analysis on the PPMI dataset due to its flexibility. It can integrate any number of modalities, whilst simultaneously allowing for the retention of the maximum number of patients possible, in contrast to other existing methodologies.…”