Cancer stemness properties, including self-renewal and differentiation, was initially attributed to normal stem cells (Malta et al., 2018). Cancer stem cells (CSCs) are responsible for cancer treatment resistance, leading to relapse, disease progression, and eventually systemic disease (Leão et al., 2017). Recent advances in high-throughput technology and machine learning have improve novel understanding of tumor heterogeneity and developed transcriptional classifications of LUAD, including molecular subtypes (Ruiz-Cordero and Devine, 2020; Wadowska et al., 2020) and immune classification (Thorsson et al., 2018). Malta et al. (2018) used an innovative one-class logistic regression machine-learning algorithm (OCLR) to analyze the molecular profiles of normal stem cell types, and they applied the OCLR-based signatures to The Cancer Genome Atlas (TCGA) datasets to obtain mRNA stemness indices (mRNAsi). Each patient of TCGA has a score of stemness index, which ranges from low (zero) to high (one) stemness. Previously, we characterized the expression of LUAD stem cell genes by mRNAsi and weighted gene co-expression network analysis (WGCNA) was used to construct a LUAD stem cell gene network (Zhang et al., 2020).
