期刊
PROTEOMICS
卷 5, 期 16, 页码 4274-4286出版社
WILEY
DOI: 10.1002/pmic.200401286
关键词
2D-DIGE; lymphoid neoplasm; multivariate analysis; statistical-learning method
To identify proteins associated with the histological subtypes of lymphoid neoplasms, we studied the proteomes of 42 cell lines from human lymphoid neoplasms including Hodgkin's lymphoma (HL; four cell lines), B cell malignancies (19 cell lines), T cell malignancies (16 cell lines), and natural killer (NK) cell lymphoma (three cell lines). The protein spots were sequentially selected by (i) Wilcoxon or Kruskal-Wallis tests to find the spots whose intensity was significantly (P <0.05) different among the cell line groups, (ii) by statistical-learning methods to prioritize the spots according to their contribution to the classification, and (iii) by unsupervised classification methods to validate the classification robustness by the selected spots. The selected spots discriminated (i) between HL cells and other cells, (ii) between the cells from B cell malignancies, T cell malignancies, and N K cell lymphoma cells, and (iii) between H L cells and anaplastic large cell lymphoma cells. Among the 31 informative protein spots, MS identified 24 proteins corresponding to 23 spots. Previous reports did not correlate these proteins to lymphocyte differentiation, suggesting that a proteomic study would identify the novel mechanisms responsible for the histogenesis of lymphoid neoplasms. These proteins may have potential as differential diagnostic markers for lymphoid neoplasms.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据