According to the serious situation of lung cancer in Chinese cancer incidence and mortality, better prognosis and early diagnosis are the key problems. These works are around of lung cancer genetic profiling, pathway signaling and tumor evolution, targeted therapy and transplant monitoring, and finally to search a better way of early diagnosis for lung cancer prevention. The study of the first large-scale sequencing effort on lung adenocarcinoma in Asian patients provides a comprehensive mutational landscape for both primary and metastatic tumours. This may thus form a basis for personalized medical care and enhance the molecular pathogenesis of metastatic lung adenocarcinoma. With findings of genomic and transcriptomic analysis in primary lung adenocarcinomas and corresponding lymph node metastases from Chinese patientsVIdentification of 13 significantly mutated genes including TP53, RHPN2, GLI3 and MRC2; TP53 mutations are significantly enriched in tumours from metastases patients; High expression level of IQGAP3 which was identified as a marker for poor prognosis. Also, next-generation sequencing is a promising cost effective and rapid technique for routine diagnostics of SNVs, InDels, CNVs, and SV in 145 genes with FFPE clinical specimens. While, couple with relatively geological distributions of different subtypes, tumor microenvironment might contribute more to genetic instability and thus tumor evolutions. As for the therapy and rejection monitoring of lung cancer patients, cell-free donor DNA had been tested. The level of donor DNA from lung allograft recipient plasma diagnosed with clinical signs of accuracy 87.03±5.9%, sensitivity 80% and specificity 100%. Besides, the same whole genome sequencing data was also effective for pathogenic factor, especially for CMV infection. DNA mutation and clonal expansion in human blood was prevalent in cancer patients and cf-DNA somatic mutation seems to be ideal cancer early diagnosis biomarkers due to its advantages. By using cell-free tumor DNA and peripheral nodule ultra-deep sequencing (>10,000 fold), mutations from nodule tissues had well performance of malignant and benign with clinic pathology; while plasma DNA mutations may have been used to predict tumor mutation level by the tumor burden model and helps molecular pathological identification in early stage.