MODILM: Towards Better Complex Diseases Classification Using a Novel Multi-omics Data Integration Learning Model

Abstract

Background Complex disease classification is an important part of the complex disease diagnosis and personalized treatment process. It has been shown that the integration of multi-omics data can analyze and classify complex diseases more accurately, because multi-omics data are highly correlated with the onset and progression of various diseases and can provide comprehensive and complementary information about a disease. However, multi-omics data of complex diseases are usually characterized by high imbalance, scale variation, high data heterogeneity and high noise interference, which pose great challenges to multi-omics integration methods. Results We propose a novel multi-omics data integration learning model called MODILM, to obtain more important and complementary information for complex disease classification from multiple omics data. Specifically, MODILM first initially constructs a similarity network for each omics data using cosine similarity measure, then learns the sample-specific features and intra-association features of single-omics data from the similarity networks using Graph Attention Networks, then maps them uniformly to a new feature space to further strengthen and extract high-level omics-specific features of the omics data using Multilayer Perceptron networks. MODILM then uses a View Correlation Discovery Network to fuse the high-level omics-specific features extracted from each omics data and further learn cross-omics features in the label space, providing unique class-level distinctiveness to classify complex diseases. We conducted extensive experiments on six benchmark datasets having the miRNA expression data, mRNA and DNA methylation data to demonstrate the superiority of our MODILM. The experimental results show that MODILM outperforms state-of-the-art methods, effectively improving the accuracy of complex disease classification. Conclusions Our MODILM provides a more competitive way to extract and integrate important and complementary information from multiple omics data, providing a very promising tool for supporting decision making for clinical diagnosis.

TEMINET: A Co-Informative and Trustworthy Multi-Omics Integration Network for Diagnostic Prediction

“…Fourteen computational approaches are investigated, which include six classic classifiers employing early integration strategies: K-nearest neighbor classifier (KNN) [30], support vector machine (SVM) [31], Lasso [32], random forest (RF) [33], XGboost [34], and fully connected neural networks (NN) [35]. Furthermore, five multi-omics classifiers are analyzed: group-regularized ridge regression [36], BPLSDA for projecting data into latent structures with discriminant analysis [37], block PLSDA with additional sparse constraints (BSPLSDA) [37], concatenation of final representations (CF) [38] for late-stage multi-omics data, gated multimodal fusion (GMU) [39] that integrates intermediate representations, along with three advanced methods, Mogonet [19], MODILM [40], and Dynamic [18]. select and utilize the informative modalities, thus ensuring a more precise characterization of each subject.…”

Intrinsic-Dimension analysis for guiding dimensionality reduction and data-fusion in multi-omics data processing

“…Note that some literature studies using TCGA datasets for testing classification models [26,27,28,29,30] already exist. However, these studies typically restrict their analysis to a maximum of four TCGA datasets, without providing clear justification for their choices.…”

404‐error “Disease not found”: Unleashing the translational potential of ‐omics approaches beyond traditional disease classification in heart failure research

“…In this context, a reductionist standpoint makes it challenging to perform differential diagnoses on pathologies with similar clinical characteristics. Recent advances in biological and medical data sources, such as ‐omics, have called into question an oversimplified view of complex biological processes that fails to account for the interconnected nature of various body systems 2–7 . Emerging technologies and the growing field of precision medicine are paving the way for a shift away from this traditional perspective.…”