Yichao Fan - Academia.edu (original) (raw)
Uploads
Papers by Yichao Fan
Journal of The American Statistical Association, 2008
Estimation of longitudinal data covariance structure poses significant challenges because the dat... more Estimation of longitudinal data covariance structure poses significant challenges because the data are usually collected at irregular time points. A viable semiparametric model for covariance matrices was proposed in Fan, that allows one to estimate the variance function nonparametrically and to estimate the correlation function parametrically via aggregating information from irregular and sparse data points within each subject. However, the asymptotic properties of their quasi-maximum likelihood estimator (QMLE) of parameters in the covariance model are largely unknown. In the current work, we address this problem in the context of more general models for the conditional mean function including parametric, nonparametric, or semiparametric. We also consider the possibility of rough mean regression function and introduce the difference-based method to reduce biases in the context of varying-coefficient partially linear mean regression models. This provides a more robust estimator of the covariance function under a wider range of situations. Under some technical conditions, consistency and asymptotic normality are obtained for the QMLE of the parameters in the correlation function. Simulation studies and a real data example are used to illustrate the proposed approach.
Journal of Urology, 2010
Identifying tumor suppressor genes silenced by promoter CpG methylation uncovers mechanisms of tu... more Identifying tumor suppressor genes silenced by promoter CpG methylation uncovers mechanisms of tumorigenesis and identifies new epigenetic biomarkers for early cancer detection. DLEC1 is located at 3p22.3, a critical tumor suppressor gene locus for renal cell carcinoma. We explored its epigenetic alteration in renal cell carcinoma and possible clinicopathological association. Materials and Methods: We examined DLEC1 expression and methylation by semiquantitative reverse transcriptase and methylation specific polymerase chain reaction in 9 renal cell carcinoma cell lines and 81 primary tumors. We also analyzed the relationship between DLEC1 methylation and clinicopathological features in patients with renal cell carcinoma. We assessed DLEC1 inhibition of renal cell carcinoma cell growth by colony formation assay. Results: DLEC1 methylation and down-regulation were detected in all renal cell carcinoma cell lines. Treatment with 5-aza-2=-deoxycytidine (Sigma®) and/or trichostatin A (Cayman Chemical, Ann Arbor, Michigan) reversed methylation and restored DLEC1 expression, indicating that methylation directly mediates its silencing. Aberrant methylation was further detected in 25 of 81 primary tumors (31%) but only 1 of 53 nonmalignant renal tissues (2%) showed methylation. DLEC1 methylation status was significantly associated with TNM classification and grade in patients with renal cell carcinoma (chi-square test p ϭ 0.01 and 0.04, respectively). DLEC1 ectopic expression in silenced renal cell carcinoma cells resulted in substantial tumor cell clonogenicity inhibition. Conclusions: To our knowledge we report for the first time that DLEC1 is often down-regulated by CpG methylation and shows tumor inhibitory function in renal cell carcinoma cells, indicating its role as a tumor suppressor. DLEC1 tumor specific methylation may serve as a biomarker for early detection and prognosis prediction of this tumor.
American Journal of Pathology, 2010
The malignant Hodgkin/Reed-Sternberg (HRS) cells of Hodgkin lymphoma (HL) are believed to derive ... more The malignant Hodgkin/Reed-Sternberg (HRS) cells of Hodgkin lymphoma (HL) are believed to derive from germinal center (GC) B cells , but lack expression of a functional B cell receptor. As apoptosis is the normal fate of B-cell receptor-negative GC B cells, mechanisms that abrogate apoptosis are thus critical in HL development , such as epigenetic disruption of certain pro-apoptotic cancer genes including tumor suppressor genes. Identifying methylated genes elucidates oncogenic mechanisms and provides valuable biomarkers; therefore, we performed a chemical epigenetic screening for methylated genes in HL through pharmacological demethylation and expression profiling. IGSF4/CADM1/TSLC1, a pro-apoptotic cell adhesion molecule of the immunoglobulin superfamily, was identified together with other methylated targets. In contrast to its expression in normal GC B cells, IGSF4 was down-regulated and methylated in HL cell lines, most primary HL, and microdissected HRS cells of 3/5 cases, but not in normal peripheral blood mononuclear cells and seldom in normal lymph nodes. We also detected IGSF4 methylation in sera of 14/18 (78%) HL patients but seldom in normal sera. Ectopic IGSF4 expression decreased HL cells survival and increased their sensitivity to apoptosis. IGSF4 induction that normally follows heat shock stress treatment was also abrogated in methylated lymphoma cells. Thus, our data demonstrate that IGSF4 silencing by CpG methylation provides an anti-apoptotic signal to HRS cells important
Journal of The American Statistical Association, 2008
Estimation of longitudinal data covariance structure poses significant challenges because the dat... more Estimation of longitudinal data covariance structure poses significant challenges because the data are usually collected at irregular time points. A viable semiparametric model for covariance matrices was proposed in Fan, that allows one to estimate the variance function nonparametrically and to estimate the correlation function parametrically via aggregating information from irregular and sparse data points within each subject. However, the asymptotic properties of their quasi-maximum likelihood estimator (QMLE) of parameters in the covariance model are largely unknown. In the current work, we address this problem in the context of more general models for the conditional mean function including parametric, nonparametric, or semiparametric. We also consider the possibility of rough mean regression function and introduce the difference-based method to reduce biases in the context of varying-coefficient partially linear mean regression models. This provides a more robust estimator of the covariance function under a wider range of situations. Under some technical conditions, consistency and asymptotic normality are obtained for the QMLE of the parameters in the correlation function. Simulation studies and a real data example are used to illustrate the proposed approach.
Journal of Urology, 2010
Identifying tumor suppressor genes silenced by promoter CpG methylation uncovers mechanisms of tu... more Identifying tumor suppressor genes silenced by promoter CpG methylation uncovers mechanisms of tumorigenesis and identifies new epigenetic biomarkers for early cancer detection. DLEC1 is located at 3p22.3, a critical tumor suppressor gene locus for renal cell carcinoma. We explored its epigenetic alteration in renal cell carcinoma and possible clinicopathological association. Materials and Methods: We examined DLEC1 expression and methylation by semiquantitative reverse transcriptase and methylation specific polymerase chain reaction in 9 renal cell carcinoma cell lines and 81 primary tumors. We also analyzed the relationship between DLEC1 methylation and clinicopathological features in patients with renal cell carcinoma. We assessed DLEC1 inhibition of renal cell carcinoma cell growth by colony formation assay. Results: DLEC1 methylation and down-regulation were detected in all renal cell carcinoma cell lines. Treatment with 5-aza-2=-deoxycytidine (Sigma®) and/or trichostatin A (Cayman Chemical, Ann Arbor, Michigan) reversed methylation and restored DLEC1 expression, indicating that methylation directly mediates its silencing. Aberrant methylation was further detected in 25 of 81 primary tumors (31%) but only 1 of 53 nonmalignant renal tissues (2%) showed methylation. DLEC1 methylation status was significantly associated with TNM classification and grade in patients with renal cell carcinoma (chi-square test p ϭ 0.01 and 0.04, respectively). DLEC1 ectopic expression in silenced renal cell carcinoma cells resulted in substantial tumor cell clonogenicity inhibition. Conclusions: To our knowledge we report for the first time that DLEC1 is often down-regulated by CpG methylation and shows tumor inhibitory function in renal cell carcinoma cells, indicating its role as a tumor suppressor. DLEC1 tumor specific methylation may serve as a biomarker for early detection and prognosis prediction of this tumor.
American Journal of Pathology, 2010
The malignant Hodgkin/Reed-Sternberg (HRS) cells of Hodgkin lymphoma (HL) are believed to derive ... more The malignant Hodgkin/Reed-Sternberg (HRS) cells of Hodgkin lymphoma (HL) are believed to derive from germinal center (GC) B cells , but lack expression of a functional B cell receptor. As apoptosis is the normal fate of B-cell receptor-negative GC B cells, mechanisms that abrogate apoptosis are thus critical in HL development , such as epigenetic disruption of certain pro-apoptotic cancer genes including tumor suppressor genes. Identifying methylated genes elucidates oncogenic mechanisms and provides valuable biomarkers; therefore, we performed a chemical epigenetic screening for methylated genes in HL through pharmacological demethylation and expression profiling. IGSF4/CADM1/TSLC1, a pro-apoptotic cell adhesion molecule of the immunoglobulin superfamily, was identified together with other methylated targets. In contrast to its expression in normal GC B cells, IGSF4 was down-regulated and methylated in HL cell lines, most primary HL, and microdissected HRS cells of 3/5 cases, but not in normal peripheral blood mononuclear cells and seldom in normal lymph nodes. We also detected IGSF4 methylation in sera of 14/18 (78%) HL patients but seldom in normal sera. Ectopic IGSF4 expression decreased HL cells survival and increased their sensitivity to apoptosis. IGSF4 induction that normally follows heat shock stress treatment was also abrogated in methylated lymphoma cells. Thus, our data demonstrate that IGSF4 silencing by CpG methylation provides an anti-apoptotic signal to HRS cells important