Table 1.
Comparisons of studies integrating omics data for prognostic predictions.
| Title | Cancer Type | Sample Size | Omics Data | Prediction Type | Methods | Reference |
|---|---|---|---|---|---|---|
| Predicting clinical outcomes from large-scale cancer genomic profiles with deep survival models | Pan-glioma (LGG/GBM), BRCA, KIPAN | Clinical and molecular data from TCGA | Gene expressions from TCGA | Survival analysis, established deep survival models to improve prognostic accuracy | Deep learning (DL) and Bayesian optimization methods | Yousefi et al., September 2017 [7] |
| Deep learning-based multi-omics integration robustly predicts survival in liver cancer | HCC | 360 patients from TCGA | RNA sequencing, miRNA sequencing, methylation data (TCGA) | Survival prognostic predictions | DL-based model, validated on five external datasets | Chaudhary et al., March 2018 [8] |
| Deep learning-based multi-omics data integration reveals two prognostic subtypes in high-risk neuroblastoma | Neuroblastoma | 407 patients from TARGET, 498 patients from SEQC | Gene expression, copy number alterations (TARGET and SEQC) | Identified two subtypes with significant survival differences | DL-based model, validated in two independent datasets | Li Zhang et al., October 2018 [9] |
| Integrative network analysis of TCGA data for ovarian cancer | Ovarian cancer | 1214 Patients from TCGA | Gene expression, methylation data, miRNA, copy number alterations (TCGA) | Predicted clinical outcomes and elucidated interplay between different levels | A new graph-based framework | Zhang et al., December 2014 [10] |
| Similarity network fusion for aggregating data types on a genomic scale | GBM, BIC, KRCCC, LSCC, COAD | Patients ranging from 92 to 215 depended on cancer type profiled by TCGA | mRNA expression, DNA methylation, miRNA expression data (TCGA) | Prediction of patients’ survival risk analysis | Similarity network fusion | Wang et al., January 2014 [11] |
TCGA, the Cancer Genome Atlas; miRNA, microRNA; TARGET, therapeutically applicable research to generate effective treatments; SEQC, sequencing quality control; LGG, low grade glioma; GBM, glioblastoma multiforme; BRCA, breast cancer; KIPAN, pan-kidney; HCC, hepatocellular carcinoma; BIC, breast invasive carcinoma; KRCCC, kidney renal clear cell carcinoma; LSCC, lung squamous cell carcinoma; COAD, colon adenocarcinoma; mRNA, messenger RNA.