Meta-Merging the Transcriptomes of Gastric Tumors Redefines the Connections among Molecular and Clinical Subtypes

Sekar, Balaji T; Ganesan, Kumaresan

doi:10.31557/APJCP.2026.27.3.883

Meta-Merging the Transcriptomes of Gastric Tumors Redefines the Connections among Molecular and Clinical Subtypes

Document Type : Research Articles

Authors

Unit of Excellence in Cancer Genetics, Department of Genetics, Centre for Excellence in Genomic Sciences, School of Biological Sciences, Madurai Kamaraj University, Madurai, India.

10.31557/APJCP.2026.27.3.883

Abstract

Introduction: The availability of a large number of cancer expression profiles presents an excellent opportunity to re-investigate various biological and clinical questions. While several expression profiles have been established for different cancers, merging them may provide a more powerful platform for extensively extrapolating molecular and clinical features across multiple cohorts. Materials and Methods: In this study, five gastric tumor expression profiles from the Gene Expression Omnibus [GEO] and one in-house cohort comprising a total of 1,060 samples were merged. The batch effect was removed using non-parametric ComBat analysis, and the seamless merging of datasets was confirmed through various parameters. Results: Extrapolation of ACRG [Asian Cancer Research Group] and TCGA [The Cancer Genome Atlas] molecular subtypes in the merged cohort of 1,060 gastric tumors revealed nine distinct clusters. Notably, the following patterns were observed: [i] mutual exclusivity between Epithelial to Mesenchymal Transition [EMT] and Microsatellite Instability [MSI] subtypes in 90% of tumors; [ii] overlapping occurrence of EMT and MSI subtypes in the remaining tumors; [iii] overlap between MSI and Epstein-Barr Virus [EBV] subtype tumors; [iv] both commonalities and differences between EMT and Genomically Stable [GS] subtypes; and [v] an association between EBV positivity and PI3K mutation. Conclusion: The current study demonstrates that compiling a larger expression profile is valuable for revisiting the molecular features and epidemiology associated with molecular subtypes, thereby aiding in the development of novel diagnostics and targeted therapeutics.

Keywords

Main Subjects