Dataset Information

Deep Learning for Integrated Analysis of Insulin Resistance with Multi-Omics Data.

ABSTRACT: Technological advances in next-generation sequencing (NGS) have made it possible to uncover extensive and dynamic alterations in diverse molecular components and biological pathways across healthy and diseased conditions. Large amounts of multi-omics data originating from emerging NGS experiments require feature engineering, which is a crucial step in the process of predictive modeling. The underlying relationship among multi-omics features in terms of insulin resistance is not well understood. In this study, using the multi-omics data of type II diabetes from the Integrative Human Microbiome Project, from 10,783 features, we conducted a data analytic approach to elucidate the relationship between insulin resistance and multi-omics features, including microbiome data. To better explain the impact of microbiome features on insulin classification, we used a developed deep neural network interpretation algorithm for each microbiome feature's contribution to the discriminative model output in the samples.

SUBMITTER: Huang E

PROVIDER: S-EPMC7918166 | biostudies-literature | 2021 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep Learning for Integrated Analysis of Insulin Resistance with Multi-Omics Data.

Huang Eunchong E Kim Sarah S Ahn TaeJin T

Journal of personalized medicine 20210215 2

Technological advances in next-generation sequencing (NGS) have made it possible to uncover extensive and dynamic alterations in diverse molecular components and biological pathways across healthy and diseased conditions. Large amounts of multi-omics data originating from emerging NGS experiments require feature engineering, which is a crucial step in the process of predictive modeling. The underlying relationship among multi-omics features in terms of insulin resistance is not well understood. ...[more]

PMID: 33671853

Similar Datasets

Project description:BackgroundBreast cancer is the most prevalent and among the most deadly cancers in females. Patients with breast cancer have highly variable survival lengths, indicating a need to identify prognostic biomarkers for personalized diagnosis and treatment. With the development of new technologies such as next-generation sequencing, multi-omics information are becoming available for a more thorough evaluation of a patient's condition. In this study, we aim to improve breast cancer overall survival prediction by integrating multi-omics data (e.g., gene expression, DNA methylation, miRNA expression, and copy number variations (CNVs)).MethodsMotivated by multi-view learning, we propose a novel strategy to integrate multi-omics data for breast cancer survival prediction by applying complementary and consensus principles. The complementary principle assumes each -omics data contains modality-unique information. To preserve such information, we develop a concatenation autoencoder (ConcatAE) that concatenates the hidden features learned from each modality for integration. The consensus principle assumes that the disagreements among modalities upper bound the model errors. To get rid of the noises or discrepancies among modalities, we develop a cross-modality autoencoder (CrossAE) to maximize the agreement among modalities to achieve a modality-invariant representation. We first validate the effectiveness of our proposed models on the MNIST simulated data. We then apply these models to the TCCA breast cancer multi-omics data for overall survival prediction.ResultsFor breast cancer overall survival prediction, the integration of DNA methylation and miRNA expression achieves the best overall performance of 0.641?±?0.031 with ConcatAE, and 0.63?±?0.081 with CrossAE. Both strategies outperform baseline single-modality models using only DNA methylation (0.583?±?0.058) or miRNA expression (0.616?±?0.057).ConclusionsIn conclusion, we achieve improved overall survival prediction performance by utilizing either the complementary or consensus information among multi-omics data. The proposed ConcatAE and CrossAE models can inspire future deep representation-based multi-omics integration techniques. We believe these novel multi-omics integration models can benefit the personalized diagnosis and treatment of breast cancer patients.

Project description:BackgroundGlycosylation plays an essential role in driving the progression and treatment resistance of hepatocellular carcinoma (HCC). However, its function in regulating the acquisition and maintenance of the cancer stemness-like phenotype in HCC remains largely unknown. There is also very little known about how CAD and other potential glycosylation regulators may influence stemness. This study explores the relationship between glycosylation and stemness in HCC.MethodsGene set variance analysis (GSVA) was used to assess the TCGA pan-cancer enrichment in glycosylation-related pathways. Univariate, LASSO, and multivariate COX regression were then used to identify prognostic genes in the TCGA-LIHC and construct a prognostic signature. HCC patients were classified into high- and low-risk subgroups based on the signature. The relationship between gene expression profiles and stemness was confirmed using bulk and single-cell RNA-sequencing data. The role of CAD and other genes in regulating the stemness of HCC was also validated by RT-qPCR, CCK-8, and colony formation assay. Copy number variation (CNV), immune infiltration, and clinical features were further analyzed in different subgroups and subsequent gene expression profiles. Sensitive drugs were also screened.ResultsIn the pan-cancer analysis, HCC was shown to have specific glycosylation alterations. Five genes, CAD, SLC51B, LGALS3, B3GAT3, and MT3, identified from 572 glycosylation-related genes, were used to construct a gene signature and predict HCC patient survival in the TCGA cohort. The results demonstrated a significant positive correlation between patients in the high-risk group and both elevated gene expression and HCC dedifferentiation status. A significant reduction in the stemness-related markers, CD24, CD44, CD20, FOXM1, and EpCAM, was found after the knockdown of CAD and other genes in HepG2 and Huh7 cells. Frequent mutations increased CNVs, immune-suppressive responses, and poor prognosis were also associated with the high-risk profile. The ICGC-LIRI-JP cohort confirmed a similar relationship between glycosylation-related subtypes and stemness. Finally, 84 sensitive drugs were screened for abnormal glycosylation of HCC, and carfilzomib was most highly correlated with CAD.ConclusionsGlycosylation-related molecular subtypes are associated with HCC stemness and disease prognosis. These results provide new directions for further research on the relationship between glycosylation and stemness phenotypes.

Dataset Information

Deep Learning for Integrated Analysis of Insulin Resistance with Multi-Omics Data.

Publications

Deep Learning for Integrated Analysis of Insulin Resistance with Multi-Omics Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure