Non-Small-Cell Lung Cancer Gene Expression Network Inference Using Gaussian Graphic Models

摘要

The Cancer Genome Atlas (TCGA) provides a rich resource that can be utilized to understand how genes interact in cancer cells. Gaussian graphical model (GGM) is often used to learn genetic networks as it defines an undirected graphical structure revealing the conditional dependences of genes. In this paper, we utilize the non-small-cell lung cancer (NSCLC) pathway from Kyoto Encyclopedia of Genes and Genomes to identify NSCLC related genes. The TCGA RNA-seq expression levels of these genes were extracted; they were cleaned, formatted and analyzed using GGM with graphical lasso. Three gene expression networks were constructed, two based on gene expression levels in cancerous lung tissues and one based on the expression levels in normal tissues. The inferred networks were examined and compared. Conserved expression-level interactions that are unique to cancerous tissues were identified.