Prediction of Lung Adenocarcinoma Driver Genes Through Protein-Protein Interaction Networks Utilizing GenePlexus

Fei Yuan,Yu-Hang Zhang,FeiMing Huang,Xiaoyu Cao,Lei Chen,JiaBo Li,WenFeng Shen,KaiYan Feng,YuSheng Bao,Tao Huang,Yu-Dong Cai
DOI: https://doi.org/10.1002/pmic.202400296
2024-12-18
PROTEOMICS
Abstract:Lung adenocarcinoma, a subtype of lung cancer, is produced by uncontrolled proliferation of somatic cells affected by some tumorigenic factors. The origin of this disease can be attributed to the concept of "cancer driver," which links the occurrence of tumor with specific changes in some key genes. These key genes can be identified at various molecular levels. Our innovative method uses a groundbreaking computing technology called GenePlexus to mine new genes related to lung adenocarcinoma. Initially, a vast network was synthesized from protein-protein interactions. Utilizing GenePlexus, we traversed paths interlinking aberrant genes across different layers and pinpointed emerging candidate genes situated on these trajectories. Finally, the candidate genes that were obtained underwent a series of filtering processes, including a permutation test, interaction test, and enrichment test. Compared with the shortest path method, GenePlexus has identified previously neglected genes involved in lung adenocarcinoma. For example, genes such as EGR2, EPHA3, FGFR4, HOXB1, and HEY1 play key roles at multiple molecular levels, including methylation, microRNA, mRNA and mutation, which affect tumorigenesis and lung cancer progression. These genes regulate various processes, from gene expression and cell proliferation to drug resistance to therapeutic drugs and the progress of lung adenocarcinoma.
What problem does this paper attempt to address?