【摘要】 目的 不同类型的癌症之间，甚至是同一癌症类型的不同患者个体之间，在代谢重编程的精确表现上都具有异质性，因此导致相同的药物在不同的癌症患者体内会产生不同的结果，以致于很难找到一种通用的治疗方案。为了探究癌症的这种异质性，本文提出了一种分析癌症个性化代谢的新方法。方法 本文使用来自癌症基因组图谱(The Cancer Genome Atlas, TCGA)的转录组数据，采用基于差异表达基因的代谢网络重构方法，通过对比每名患者的癌细胞和邻近正常细胞的基因表达数据之间的差异，还原重构出癌细胞相对原发组织细胞特有的那部分代谢网络。通过模拟基因敲除等手段，寻找潜在的癌症个性化治疗靶点基因。结果 本文对来自TCGA的17种癌症类型的679对样本进行分析，发现了它们在癌症特异性代谢网络上的巨大差异，并得到大量的个性化治疗靶点。特别是其中63个靶点基因，它们只对特定的某位患者有效。结论 不同的患者细胞癌变后发生的代谢变化具有很大的差异，需要进行个性化治疗方案的研究。本文提出的分析方法为癌症的个性化分析提供了新的思路，找到的个性化靶点也可以进行进一步的临床研究。更多还原

【Abstract】 Objective There is heterogeneity in the precise performance of metabolic reprogramming between different types of cancer, even between different patients of the same cancer type. Therefore, the same drug can produce different results in different cancer patients, so that it is difficult to find a general treatment plan. In order to explore this heterogeneity of cancer, this article proposes a new method to analyze the personalized metabolism of cancer. Methods This article uses transcriptome data from The Cancer Genome Atlas(TCGA),and uses a metabolic network reconstruction method based on differentially expressed genes. By comparing the difference of the gene expression data between cancer cells and neighboring normal cells in each patient, the metabolic network that is unique to the cancer cell relative to the original tissue cells can be restored and reconstructed. By simulating gene knockout and other means, we can find potential target genes for personalized cancer therapy. Results This article analyzes 679 samples from 17 cancer types from TCGA and finds huge differences in cancer-specific metabolic networks, and obtains a large number of personalized treatment targets. In particular, 63 of these target genes are only effective for a specific patient. Conclusions The metabolic abnormalities that occur after different individuals’ cells become cancerous are very different, and it is necessary to conduct research on individualized treatment plans. The analysis method proposed in this paper provides new ideas for the personalized analysis of cancer, and the personalized targets found can also be used for further clinical research.更多还原