Integrating multi-omics data through deep learning for accurate cancer prognosis prediction

Abstract

Motivation: Accurately predicting cancer prognosis is necessary to choose precise strategies of treatment for patients. One of effective approaches in the prediction is the integration of multi-omics data, which reduces the impact of noise within single omics data. However, integrating multi-omics data brings large number of redundant variables and relative small sample sizes. In this study, we employed Autoencoder networks to extract important features that were then input to the proportional hazards model to predict the cancer prognosis. Results: The method was applied to 12 common cancers from the Cancer Genome Atlas. The results show that the multi-omics averagely improves 4.1% C-index for prognosis prediction over single mRNA data, and our method outperforms previous approaches by at least 7.4%. A comparison of the contribution of single omics data show that mRNA contributes the most, followed by the DNA methylation, miRNA, and the copy number variation. In the case study for differential gene expression analysis, we identified 161 differentially expressed genes in the cervical cancer, among which 77 genes (65.8%) have been proven to be associated with cancer. In addition, we performed the cross-cancer test where the model trained on one cancer was used to predict the prognosis of another cancer, and found 23 pairs of cancers have a C-index larger than 0.5, with the largest value of 0.68. Thus, this study has provided a deep learning framework to effectively integrate multiple omics data to predict cancer prognosis.

Integrative Network Fusion: A Multi-Omics Approach in Molecular Profiling

“…Overall, INF improves over the performance of single layers and naive juxtaposition on all four oncogenomics tasks, extracting a biologically meaningful compact set of predictive biomarkers. Notably, the transcriptomics layer is prevalent inside the inferred INF signatures, consistently with published findings ( 9 ).…”

“…Indeed, the underlying hypothesis of multi-omics integration is that different omics data can provide complementary information ( 56 ) [although sometimes redundant ( 9 )], and thus a broader insight with respect to single-layer analysis, for a better understanding of disease mechanisms ( 59 ). This assumption has been confirmed by multiple studies on diverse diseases, such as cardiovascular disease ( 60 ), diabetes ( 61 ), liver disease ( 62 ), or mitochondrial diseases ( 63 ), and also longitudinally ( 64 ), suggesting that the more complex the disease the more advantageous the integration.…”

“…Our investigation shows that the signatures identified through the INF framework include biological markers that are relevant in the tasks under analysis and are consistent with previously published data. Further, as in (9), the largest contribution in the biomarkers' lists is provided by gene expression, while epigenomics, proteomics and miRNA transcriptomics play a minor role.…”

Integrative Network Fusion: a multi-omics approach in molecular profiling

“…More recently, the success of deep learning algorithms in various bioinformatics fields (91) prompted the adoption of deep neural network for omics-integration in precision oncology. Autoencoders and convolutional neural networks have been effectively trained for the prediction of prognostic outcomes (92,9), response to chemotherapeutic drugs (47), and gene targeting (93), by adopting either an early-integration (9,93) or a late-integration (92,47). Although deep learning models hold the potential to include image-derived features in the integration workflow, they suffer from interpretability and generalization issues (94).…”

Leveraging Methylation Alterations to Discover Potential Causal Genes Associated with the Survival Risk of Cervical Cancer in TCGA through a Two-stage Inference Approach

“…After normalization with the trimmed mean of M values (TMM) method, differential expressed genes(DEGs) were screened via the exact test based on quantile-adjusted conditional maximum likelihood estimation[61,62] implemented in the edgeR package (version 3.30.3)[63,64]. Following previous work[65,66], DEGs were defined if FDR < 0.05 and |log 2ResultsCervical cancer datasets in TCGA and methylation-regulated genesAfter quality control we reserved 485,577 DNA methylation GpG sites and 3 clinical covariates (i.e. age of onset, clinical stage, and tumor status) up to 190 cervical cancer patients of European ancestry.…”