Screening and Identification of a Prognostic Model of Ovarian Cancer by Combination of Transcriptomic and Proteomic Data

Biomolecules. 2023 Apr 18;13(4):685. doi: 10.3390/biom13040685.

Abstract

The integration of transcriptome and proteome analysis can lead to the discovery of a myriad of biological insights into ovarian cancer. Proteome, clinical, and transcriptome data about ovarian cancer were downloaded from TCGA's database. A LASSO-Cox regression was used to uncover prognostic-related proteins and develop a new protein prognostic signature for patients with ovarian cancer to predict their prognosis. Patients were brought together in subgroups using a consensus clustering analysis of prognostic-related proteins. To further investigate the role of proteins and protein-coding genes in ovarian cancer, additional analyses were performed using multiple online databases (HPA, Sangerbox, TIMER, cBioPortal, TISCH, and CancerSEA). The final resulting prognosis factors consisted of seven protective factors (P38MAPK, RAB11, FOXO3A, AR, BETACATENIN, Sox2, and IGFRb) and two risk factors (AKT_pS473 and ERCC5), which can be used to construct a prognosis-related protein model. A significant difference in overall survival (OS), disease-free interval (DFI), disease-specific survival (DSS), and progression-free interval (PFI) curves were found in the training, testing, and whole sets when analyzing the protein-based risk score (p < 0.05). We also illustrated a wide range of functions, immune checkpoints, and tumor-infiltrating immune cells in prognosis-related protein signatures. Additionally, the protein-coding genes were significantly correlated with each other. EMTAB8107 and GSE154600 single-cell data revealed that the genes were highly expressed. Furthermore, the genes were related to tumor functional states (angiogenesis, invasion, and quiescence). We reported and validated a survivability prediction model for ovarian cancer based on prognostic-related protein signatures. A strong correlation was found between the signatures, tumor-infiltrating immune cells, and immune checkpoints. The protein-coding genes were highly expressed in single-cell RNA and bulk RNA sequencing, correlating with both each other and tumor functional states.

Keywords: ovarian cancer; proteome; single cell; transcriptome; tumor-infiltrating immune cells.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Biomarkers, Tumor / genetics
  • Early Detection of Cancer*
  • Female
  • Humans
  • Ovarian Neoplasms* / diagnosis
  • Ovarian Neoplasms* / genetics
  • Prognosis
  • Proteome / genetics
  • Proteomics
  • Transcriptome

Substances

  • Proteome
  • Biomarkers, Tumor

Grants and funding

This work was supported by grants from the Key Research and Development Program of Guangxi (2021AB13014), Major Project of Guangxi Innovation Driven (AA18118016), National Key Research and Development Program of China (2017YFC0908000), Natural Key Research and Development Project (2020YFA0113200), Natural Science Foundation of China (81770759, 82060145, 31970814), Natural Science Foundation of Guangxi Zhuang Autonomous Region (2021JJA140912), Advanced Innovation Teams and Xinghu Scholars Program of Guangxi Medical University, Guangxi Key Laboratory for Genomic and Personalized Medicine (19-050-22, 19-185-33, 20-065-33, 22-35-17), major project of Scientific Research and Technology Development Plan of Nanning (20221023), and Guangxi Natural Science Foundation (2022GXNSFAA035641).