The development of cancer diseases is driven by the accumulation of many oncogenesis-related genetic alterations and tumorigenesis is triggered by complex networks of involved genes rather than independent actions. The interactions among genetic factors are believed to play important roles in carcinogenesis and contribute to the missing heritability. In this proposal, we designed a filtered gene-gene interaction analysis aiming to identify the epistasis involving important oncogenesis-related genes. Besides the advantage of dimensional reduction and increased power, this filtered interaction analysis will help us identify novel oncogenes or tumor suppressor genes implicated in lung cancer development that will not be revealed in main effect association analysis. It will also identify novel susceptibility variants, including those in intergenic and non-coding regions, affecting lung cancer risk by interacting/modulating with oncogenesis-related genes. Lung cancer is a heterogeneous disease and researchers have identified vast differences in genomic attributes. However, the knowledge about epistatic features in lung cancer subtypes is limited. We will conduct a stratified epistasis analysis by lung cancer histology subtype in the proposed study to reveal subtype-specific genetic interactions and gene networks. The stratified analysis by histology will enhance our understanding about this complicated disease mechanism in lung cancer and has the potential to contribute to precision medicine in lung cancer treatment. A main challenge in genetic association studies is to understand the functional consequences of identified genetic variants. In this study, we proposed functional annotation analysis including eQTL gene expression analysis, pathway and gene network analysis, and functional annotation of epistasis-involved SNPs. This integrative epistasis and functional annotation analysis will help us pinpoint the causal epistasis and characterize the epistasis-involved genes or regions in lung cancer risk development. It will be a pilot study to explore how the regulatory non-coding variants impact lung cancer risk by interacting with oncogenesis-related genes. The proposed study will provide insights about the complicated biological interactions that are critical for gene regulation, biochemical networks, and developmental pathways implicated in lung carcinogenesis. In order to finish the proposed study, we collected the genotype data from three independent GWAS studies including 24,037 lung cancer patients and 20,401 healthy controls from the Caucasian population. The genotype and gene expression data in lung tissues from 409 individuals will be applied for eQTL gene expression analysis.
The interactions among genetic factors are believed to play important roles in carcinogenesis and contribute to the missing heritability. In this proposal, we designed an integrative filtered epistasis and functional annotation analysis to identify the genetic interactions involving important oncogenesis-related genes in lung cancer development. Besides the advantage of dimensional reduction and increased power, this filtered interaction analysis will help us identify novel variants, including the regulatory elements in non-coding regions, implicated in lung cancer development that will not be revealed in main effect association analysis.
