Incorporating deep learning and multi-omics autoencoding for analysis of lung adenocarcinoma prognostication

Tzong Yi Lee, Kai Yao Huang, Cheng Hsiang Chuang, Cheng Yang Lee, Tzu Hao Chang

Research output: Contribution to journalArticlepeer-review

52 Citations (Scopus)


Lung cancer is the most occurring cancer type, and its mortality rate is also the highest, among them lung adenocarcinoma (LUAD) accounts for about 40 % of lung cancer. There is an urgent need to develop a prognosis prediction model for lung adenocarcinoma. Previous LUAD prognosis studies only took single-omics data, such as mRNA or miRNA, into consideration. To this end, we proposed a deep learning-based autoencoding approach for combination of four-omics data, mRNA, miRNA, DNA methylation and copy number variations, to construct an autoencoder model, which learned representative features to differentiate the two optimal patient subgroups with a significant difference in survival (P = 4.08e-09) and good consistency index (C-index = 0.65). The multi-omics model was validated though four independent datasets, i.e. GSE81089 for mRNA (n = 198, P = 0.0083), GSE63805 for miRNA (n = 32, P = 0.018), GSE63384 for DNA methylation (n = 35, P = 0.009), and TCGA independent samples for copy number variations (n = 94, P = 0.0052). Finally, a functional analysis was performed on two survival subgroups to discover genes involved in biological processes and pathways. This is the first study incorporating deep autoencoding and four-omics data to construct a robust survival prediction model, and results show the approach is useful at predicting LUAD prognostication.

Original languageEnglish
Article number107277
JournalComputational Biology and Chemistry
Publication statusPublished - Aug 2020


  • Autoencoder
  • Deep learning
  • Lung adenocarcinoma
  • Machine learning
  • Multi-omics
  • Prognosis prediction
  • Survival analysis

ASJC Scopus subject areas

  • Structural Biology
  • Biochemistry
  • Organic Chemistry
  • Computational Mathematics


Dive into the research topics of 'Incorporating deep learning and multi-omics autoencoding for analysis of lung adenocarcinoma prognostication'. Together they form a unique fingerprint.

Cite this