Zhanfeng Wang

A parsimonious threshold-independent protein feature selection method through the area under receiver operating characteristic curve

MOTIVATION Protein expression profiling for differences indicative of early cancer holds promise for improving diagnostics. Due to their high dimensionality, statistical analysis of proteomic data from mass spectrometers is challenging in many aspects such as dimension reduction, feature subset selection as well as construction of classification rules. Search of an optimal feature subset, commonly known as the feature subset selection (FSS) problem, is an important step towards disease classification/diagnostics with biomarkers. METHODS We develop a parsimonious threshold-independent feature selection (PTIFS) method based on the concept of area under the curve (AUC) of the receiver operating characteristic (ROC). To reduce computational complexity to a manageable level, we use a sigmoid approximation to the empirical AUC as the criterion function. Starting from an anchor feature, the PTIFS method selects a feature subset through an iterative updating algorithm. Highly correlated features that have similar discriminating power are precluded from being selected simultaneously. The classification rule is then determined from the resulting feature subset. RESULTS The performance of the proposed approach is investigated by extensive simulation studies, and by applying the method to two mass spectrometry data sets of prostate cancer and of liver cancer. We compare the new approach with the threshold gradient descent regularization (TGDR) method. The results show that our method can achieve comparable performance to that of the TGDR method in terms of disease classification, but with fewer features selected. AVAILABILITY Supplementary Material and the PTIFS implementations are available at http://staff.ustc.edu.cn/~ynyang/PTIFS. SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.

Surface enhanced laser desorption/ionization profiling: New diagnostic method of HBV-related hepatocellular carcinoma

Background and Aim:  To screen for serum biomarkers of HBV‐related hepatocellular carcinoma (HCC) and HBV‐related liver cirrhosis (LC) in an attempt to seek a new method for differential diagnosis of HCC and LC using surface‐enhanced laser desorption/ionization time‐of‐flight mass spectrometry (SELDI‐TOF‐MS) techniques.

Group variable selection and estimation in the tobit censored response model

The tobit censored response model plays an important role in analyzing the dependent variable with a constraint at a pre-specified point such as 0, and is widely used in econometrics research. For this regression model, there are few studies on variable selection in a group manner. In this paper, we present a group variable selection and estimation method for predefined groups of variables. The proposed method selects variables significantly contributing to the regression model and presents consistent estimates of parameters in the selected groups. The asymptotic properties of the resulting estimates are similar to oracle properties. The performance of our method is evaluated with extensive simulation studies and a real example from a married womens work hour study.

Asymptotic Normality of Maximum Quasi-Likelihood Estimators in Generalized Linear Models with Fixed Design*

AbstractIn generalized linear models with fixed design, under the assumption

Random weighting approximation for Tobit regression models with longitudinal data

\underline \lambda _n \to \infty

A relative error estimation approach for multiplicative single index model

and other regularity conditions, the asymptotic normality of maximum quasi-likelihood estimator

Sequential estimate for generalized linear models with uncertain number of effective variables

\hat \beta _n

Approximation by randomly weighting method in censored regression model

, which is the root of the quasi-likelihood equation with natural link function

Change-point estimation for censored regression model

\sum\nolimits_{i = 1}^n {X_i \left( {y_i - \mu \left( {X_i^\prime \beta } \right)} \right) = 0}

A relative error estimation approach for single index model

, is obtained, where