基于 TARGET 转录组数据的儿童骨肉瘤预后预测与决策支持摘 要骨肉瘤是发生在 20 岁以下青少年或儿童的一种恶性骨肿瘤，也叫成骨肉瘤，是小儿骨恶性肿瘤中最多见的，约为小儿肿瘤的 5%。该疾病的显著症状是因肿瘤组织不断侵蚀和破坏骨皮质而带来的源源不断的疼痛。骨肉瘤一旦经病理证实，就应该立即开始前期的化学治疗和手术治疗。虽然放化疗具有一定效果，但是总体预后不佳，需对其分子机制开展进一步的研究。近年来，随着组学技术的不断发展，对于各种不同的生物数据获取与分析的方法日益成熟，在癌症的诊治过程中，通过对组学数据进行分析与挖掘，利用处理后的患者数据建模并且对其进行预后预测，能够对患者提供更精准的临床决策支持。基于此，本文利用来自 TARGET 数据仓库中的 88 例儿童骨肉瘤患者的转录组数据，利用生物信息学和数据分析、挖掘方法，通过对儿童转录组数据进行分析，找出正常儿童和患病儿童的差异 lncRNAs，然后对有差异的 lncRNAs 进行单因素和多因素 Cox 生存分析，建立骨肉瘤预测的 Cox 生存模型，按风险情况将患者病情划分，对患者的生存情况进行预测，并对后期医生的临床决策提供支持。关键词：骨肉瘤；临床决策支持；预后预测IAbstractOsteosarcoma is a type of malignant bone tumor that occurs in adolescents or children under the age of 20, also known as osteosarcoma. It is the most common malignant tumor of bone in children, about 5% of children's tumors. The obvious symptom of the disease is the continuous pain caused by the continuous erosion and destruction of the bone cortex by the tumor tissue. Once the osteosarcoma has been confirmed by pathology, early chemotherapy and surgery should be started immediately. Although chemoradiotherapy has certain effects, the overall prognosis is poor, and further research on its molecular mechanism is needed.In recent years, with the continuous development of omics technology, methods for acquiring and analyzing various biological data have become increasingly mature. In the process of cancer diagnosis and treatment, by analyzing and proce...
