RNA-sequencing (RNA-seq) data, DNA methylation data generated by Illumina Infinium HumanMethylation 450 BeadChip array, copy number variation (CNV) data (Masked Copy Number Segment, affymetrix snp 6.0), single nucleotide variant (SNV) data (MuTect2. Variant0. Maf), and LUAD clinical follow-up pathological data were downloaded from The Cancer Genome Atlas (TCGA) (https://portal.gdc.cancer.gov/). The expression profile data was changed from fragments per kilobase million (FPKM) to transcripts per million (TPM) and transformed into log2. The processing of DNA methylation data involved the use of the k-nearest neighbors (KNN) algorithm to complete the missing values and retrograde, the removal of cytosine-guanine (CpG) sites with cross-reflection, and the exclusion of methylation sites on X and Y chromosomes. The stemness index was derived from Malta et al. (13 (link)). messenger RNA (mRNA) expression-based stemness indices (mRNAsi) was calculated based on the expression profile data, while DNA methylation-based stemness index (mDNAsi) was calculated based on the methylation data. The immune cell biomarkers gene set was obtained from the article of Charoentong et al. (14 (link)), which includes 782 genes. The workflow of this study is summarized in Figure S1. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013).