OTCE is designed to integrate multi-omics data for predicting the prognosis risk of cancer patients. In this framework, tabular data from three modalities are first converted into image forms to preserve spatial structure and global correlations, enabling unified representation and mitigating the limitations of late-fusion strategies. A parallel feature extraction network composed of different modules is then employed for feature extraction, and the resulting representations are fed into a survival analysis network for accurate prognosis risk prediction.
python, R
In this study, we utilized cancer datasets from the TCGA portal (https://tcga-data.nci.nih.gov/tcga/). All these datasets were downloaded by using the R package “TCGA-assembler”, which contains different types of multi-omics data: mRNA, miRNA, and DNA methylation. Here, “mRNA” was RNA sequencing data generated by UNC Illumina HiSeq_RNASeq V2; Level 3, “miRNA” was miRNA sequencing data obtained by BCGSC Illumina HiSeq miRNASeq, DNA methylation data was generated by USC HumanMethylation450.
OTCE was a framework with three steps. Firstly please use the code in the link: https://github.com/ztpub/Vec2image to generate the figures with multi-omics information. After that the corresponding figures will be processed by mixvae0all1.py for prognosis prediction model construction. At last the analysis.R in the analysis folder can be used for differential expression analysis.
Beforing using please checking the file folder path.