Feature selection and machine learning with mass spectrometry data for distinguishing cancer and non-cancer samples

This is a comparative study of various clustering and classification algorithms as applied to differentiate cancer and non-cancer protein samples using mass spectrometry data. Our study demonstrates the usefulness of a feature selection step prior to applying a machine learning tool. A natural and common choice of a feature selection tool is the collection of marginal p-values obtained from t-tests for testing the intensity differences at each m/z ratio in the cancer versus non-cancer samples. We study the effect of selecting a cutoff in terms of the overall Type 1 error rate control on the performance of the clustering and classification algorithms using the significant features. For the classification problem, we also considered m/z selection using the importance measures computed by the Random Forest algorithm of Breiman. Using a data set of proteomic analysis of serum from ovarian cancer patients and serum from cancer-free individuals in the Food and Drug Administration and National Cancer Institute Clinical Proteomics Database, we undertake a comparative study of the net effect of the machine learning algorithm–feature selection tool–cutoff criteria combination on the performance as measured by an appropriate error rate measure.     

DENSITY ESTIMATION FOR CLUSTERED DATA

The commonly used survey technique of clustering introduces dependence into sample data. Such data is frequently used in economic analysis, though the dependence induced by the sample structure of the data is often ignored. In this paper, the effect of clustering on the non-parametric, kernel estimate of the density, f(x), is examined. The window width commonly used for density estimation for the case of i.i.d. data is shown to no longer be optimal. A new optimal bandwidth using a higher-order kernel is proposed and is shown to give a smaller integrated mean squared error than two window widths which are widely used for the case of i.i.d. data. Several illustrations from simulation are provided.     

Statistical Analysis of Marginal Count Failure Data

Manufacturers want to assess the quality andreliability of their products. Specifically, they want to knowthe exact number of failures from the sales transacted duringa particular month. Information available today is sometimesincomplete as many companies analyze their failure data simplycomparing sales for a total month from a particular departmentwith the total number of claims registered for that given month.This information—called marginal count data—is, thus,incomplete as it does not give the exact number of failures ofthe specific products that were sold in a particular month. Inthis paper we discuss nonparametric estimation of the mean numbersof failures for repairable products and the failure probabilitiesfor nonrepairable products. We present a nonhomogeneous Poissonprocess model for repairable products and a multinomial modeland its Poisson approximation for nonrepairable products. A numericalexample is given and a simulation is carried out to evaluatethe proposed methods of estimating failure probabilities undera number of possible situations.     

A Multiple Imputation Approach to Linear Regression with Clustered Censored Data

Weextend Wei and Tanner's (1991) multiple imputation approach insemi-parametric linear regression for univariate censored datato clustered censored data. The main idea is to iterate the followingtwo steps: 1) using the data augmentation to impute for censoredfailure times; 2) fitting a linear model with imputed completedata, which takes into consideration of clustering among failuretimes. In particular, we propose using the generalized estimatingequations (GEE) or a linear mixed-effects model to implementthe second step. Through simulation studies our proposal comparesfavorably to the independence approach (Lee et al., 1993), whichignores the within-cluster correlation in estimating the regressioncoefficient. Our proposal is easy to implement by using existingsoftwares.     

Bayesian inference for generalized additive mixed models based on Markov random field priors

Most regression problems in practice require flexible semiparametric forms of the predictor for modelling the dependence of responses on covariates. Moreover, it is often necessary to add random effects accounting for overdispersion caused by unobserved heterogeneity or for correlation in longitudinal or spatial data. We present a unified approach for Bayesian inference via Markov chain Monte Carlo simulation in generalized additive and semiparametric mixed models. Different types of covariates, such as the usual covariates with fixed effects, metrical covariates with non-linear effects, unstructured random effects, trend and seasonal components in longitudinal data and spatial covariates, are all treated within the same general framework by assigning appropriate Markov random field priors with different forms and degrees of smoothness. We applied the approach in several case-studies and consulting cases, showing that the methods are also computationally feasible in problems with many covariates and large data sets. In this paper, we choose two typical applications.     

Case–control studies with complex sampling

The use of complex sampling designs in population-based case–control studies is becoming more common, particularly for sampling the control population. This is prompted by all the usual cost and logistical benefits that are conferred by multistage sampling. Complex sampling has often been ignored in analysis but, with the advent of packages like SUDAAN, survey-weighted analyses that take account of the sample design can be carried out routinely. This paper explores this approach and more efficient alternatives, which can also be implemented by using readily available software.     

Overcoming biases and misconceptions in ecological studies

The aggregate data study design provides an alternative group level analysis to ecological studies in the estimation of individual level health risks. An aggregate model is derived by aggregating a plausible individual level relative rate model within groups, such that population-based disease rates are modelled as functions of individual level covariate data. We apply an aggregate data method to a series of fictitious examples from a review paper by Greenland and Robins which illustrated the problems that can arise when using the results of ecological studies to make inference about individual health risks. We use simulated data based on their examples to demonstrate that the aggregate data approach can address many of the sources of bias that are inherent in typical ecological analyses, even though the limited between-region covariate variation in these examples reduces the efficiency of the aggregate study. The aggregate method has the potential to estimate exposure effects of interest in the presence of non-linearity, confounding at individual and group levels, effect modification, classical measurement error in the exposure and non-differential misclassification in the confounder.     

Flexible regression modeling

The author proposes a new method for flexible regression modeling of multi‐dimensional data, where the regression function is approximated by a linear combination of logistic basis functions. The method is adaptive, selecting simple or more complex models as appropriate. The number, location, and (to some extent) shape of the basis functions are automatically determined from the data. The method is also affine invariant, so accuracy of the fit is not affected by rotation or scaling of the covariates. Squared error and absolute error criteria are both available for estimation. The latter provides a robust estimator of the conditional median function. Computation is relatively fast, particularly for large data sets, so the method is well suited for data mining applications.     

Bootstrapping Malmquist Indices for Danish Seiners in the North Sea and Skagerrak

In connection with assessing how an ongoing development in fisheries management may change fishing activity, evaluation of Total Factor Productivity (TFP) change over a period, including efficiency, scale and technology changes, is an important tool. The Malmquist index, based on distance functions evaluated with Data Envelopment Analysis (DEA), is often employed to estimate TFP changes. DEA is generally gaining attention for evaluating efficiency and capacity in fisheries. One main criticism of DEA is that it does not have any statistical foundation, i.e. that it is not possible to make inference about DEA scores or related parameters. The bootstrap method for estimating confidence intervals of deterministic parameters can however be applied to estimate confidence intervals for DEA scores. This method is applied in the present paper for assessing TFP changes between 1987 and 1999 for the fleet of Danish seiners operating in the North Sea and the Skagerrak.     

A flexible semiparametric transformation model for survival data

I suggest an extension of the semiparametric transformation model that specifies a time-varying regression structure for the transformation, and thus allows time-varying structure in the data. Special cases include a stratified version of the usual semiparametric transformation model. The model can be thought of as specifying a first order Taylor expansion of a completely flexible baseline. Large sample properties are derived and estimators of the asymptotic variances of the regression coefficients are given. The method is illustrated by a worked example and a small simulation study. A goodness of fit procedure for testing if the regression effects lead to a satisfactory fit is also suggested.