论文部分内容阅读
目的从分子生物学的角度对弥漫大B细胞淋巴瘤(DLBCL)基因表达样本进行分类,改进基本粒子群优化分类算法在分类精确性和稳定性方面的不足。方法给出一种基于双层粒子群优化(TLPSO)的分类算法,选取141个弥漫大B细胞淋巴瘤样本,包括氧化磷酸化(OxPhos)、B细胞受体(BCR)和宿主反应(HR)3种亚型,随机选取训练集和测试集以获取不同样本组合,与基本粒子群优化(PSO)分类算法进行比较。结果基于TLPSO的分类算法获得较好分类结果,最佳分类预测结果数和分类结果分布2项指标均优于PSO算法。结论双层粒子群优化分类算法能够对弥漫大B细胞淋巴瘤基因表达样本进行准确和稳定分类,能为临床肿瘤基因表达样本的分类定型提供依据。
Objective To classify the samples of diffuse large B cell lymphoma (DLBCL) gene expression from the perspective of molecular biology and to improve the deficiencies in the classification accuracy and stability of the basic particle swarm optimization algorithm. Methods A classification algorithm based on bi-level particle swarm optimization (TLPSO) is presented. 141 samples of diffuse large B-cell lymphoma including oxidative phosphorylation (OxPhos), B cell receptor (BCR) and host reaction (HR) Three kinds of subtypes were selected. The training set and the test set were randomly selected to obtain different combinations of samples and compared with the basic particle swarm optimization (PSO) classification algorithm. Results The classification results based on TLPSO obtained better classification results, and the two indexes of the best classification prediction result and the classification result distribution were better than the PSO algorithm. Conclusion Two-layer particle swarm optimization classification algorithm can accurately and stably classify diffuse large B-cell lymphoma gene expression samples and provide a basis for the classification and typing of clinical tumor gene expression samples.