Integrative analysis of genetic variability and functional traits in lung adenocarcinoma epithelial cells via single-cell RNA sequencing, GWAS, bayesian deconvolution, and machine learning.

Abstract

Background: Lung adenocarcinoma remains a leading cause of cancer-related mortality worldwide, characterized by high genetic and cellular heterogeneity, especially within the tumor microenvironment.

Objective: This study integrates single-cell RNA sequencing (scRNA-seq) with genome-wide association studies (GWAS) using Bayesian deconvolution and machine learning techniques to unravel the genetic and functional complexity of lung adenocarcinoma epithelial cells.

Methods: We performed scRNA-seq and GWAS analysis to identify critical cell populations affected by genetic variations. Bayesian deconvolution and machine learning techniques were applied to investigate tumor progression, prognosis, and immune-epithelial cell interactions, particularly focusing on immune checkpoint markers such as PD-L1 and CTLA-4.

Results: Our analysis highlights the importance of genes like SLC2A1, which regulates glucose metabolism and correlates with tumor invasiveness and poor prognosis. Immune-epithelial interactions suggest a suppressive tumor microenvironment, potentially hindering immune responses. Additionally, machine learning models identify core prognostic genes such as F12, GOLM1, and S100P, which are significantly associated with patient survival.

Conclusions: This comprehensive approach provides novel insights into lung adenocarcinoma biology, emphasizing the role of genetic and immune factors in tumor progression. The findings support the development of personalized therapeutic strategies targeting both tumor cells and the immune microenvironment.