Network molecular nutrition framework for exploring the oncogenic mechanisms of nutrient deficiency in cervical cancer

Background: Vitamin and mineral deficiencies have been implicated in DNA damage and increased cancer risk. However, the precise oncogenic mechanisms associated with nutrient deficiencies remain unclear. This study proposes a novel framework, Network Molecular Nutrition (NetMN), to investigate the roles of nutrients in the oncogenic mechanisms of cervical cancer. Method: The study employs EL-NetMN, an evolutionary learning-derived method, to construct the NetMN framework. NetMN comprises three key identification tasks: 1) discovering a survival-related long non-coding RNA (lncRNA) signature, 2) identifying hub genes targeted by this signature, and 3) exploring pathways and nutrients/drug associations linked to these hub genes. NetMN reveals nutrient roles in cervical cancer through lncRNA-target-pathway-nutrients/drugs network analysis. Results: EL-NetMN identified an 8-lncRNA signature and established a Cox proportional hazards model, called Cox-EL, which achieved a C-index of 0.674±0.02. This performance surpassed that of Cox-LASSO, Cox-Elastic, and the random survival forest models. Consequently, EL-NetMN identified six nutrients, vitamin A, vitamin E, lycopene, selenium, zinc, and copper, along with three established cervical cancer drugs: cisplatin, paclitaxel, and vincristine. Notably, selenium was found to influence six significant hub genes and participate in eight critical biological pathways mediated by three survival-associated lncRNAs: SNHG17, ILF3-AS1, and LINC01089. These findings highlight the potential role of nutrient deficiencies in cervical cancer progression. Conclusion: This work presents a robust and systematic NetMN framework that integrates survival-associated lncRNAs, hub genes, biological pathways, and nutrient/drug associations to identify beneficial nutrients and infer nutrient-related regulatory mechanisms in cancer development. Using cervical cancer as a representative model, NetMN highlights its potential clinical and translational relevance for understanding nutrient-associated cancer progression.