FiGS: a filter-based gene selection workbench for microarray data

Abstract

BackgroundThe selection of genes that discriminate disease classes from microarray data is widely used for the identification of diagnostic biomarkers. Although various gene selection methods are currently available and some of them have shown excellent performance, no single method can retain the best performance for all types of microarray datasets. It is desirable to use a comparative approach to find the best gene selection result after rigorous test of different methodological strategies for a given microarray dataset.ResultsFiGS is a web-based workbench that automatically compares various gene selection procedures and provides the optimal gene selection result for an input microarray dataset. FiGS builds up diverse gene selection procedures by aligning different feature selection techniques and classifiers. In addition to the highly reputed techniques, FiGS diversifies the gene selection procedures by incorporating gene clustering options in the feature selection step and different data pre-processing options in classifier training step. All candidate gene selection procedures are evaluated by the .632+ bootstrap errors and listed with their classification accuracies and selected gene sets. FiGS runs on parallelized computing nodes that capacitate heavy computations. FiGS is freely accessible at http://gexp.kaist.ac.kr/figs.ConclusionFiGS is an web-based application that automates an extensive search for the optimized gene selection analysis for a microarray dataset in a parallel computing environment. FiGS will provide both an efficient and comprehensive means of acquiring optimal gene sets that discriminate disease states from microarray datasets.

Selección y clasificación de genes cancerígenos utilizando un método híbrido filtro/wrapper

“…La alta dimensión del microarreglo genera un problema de precisión de análisis y complejidad computacional [5]. Los microarreglos no solo cuentan con una gran cantidad de atributos (genes) y un número limitado de muestras, también contienen dos o más número de clases (categorías) a las que pertenece cada uno de los atributos, además miles de los genes son redundantes o ruidosos [6].…”

Gene Ranking Techniques via Attribute Evaluation Algorithms for DNA Microarray Analysis

“…[5,9], third: the evaluation by dependency measure (correlation coefficient CC) [8], fourth: the evaluation by consistency (min-features bias). There are other filter-base attributes selection method such as, Markov blanket-embedded genetic algorithm for gene selection [10], Chi-square, Relief-F symmetric uncertainty [3], Signal-to-Noise ratio (SNR) [8], t-statistics (TS) [4] One-gene-ata-time filter methods, such as ranking [11], Wilcoxon rank sum test [12]. In the other hand, a wrapper method embeds a gene selection method within a classifier, as shown in figure 2, for instance of a wrapper method is SVMs [13,14].…”

“…SVMs use the recursive attribute elimination (RFE) approach to eliminate the attributes iteratively in a greedy approach until the largest margin of partition is reached. Shortly In spite of existing 32 different attribute selection methods yet, no single gene selection method can generally improve the performance of classification algorithms in terms of efficiency and accuracy thus there are about 60 different gene selection procedures developed by combining the attribute selection methods [12].…”

Improving PLS–RFE based gene selection for microarray data classification

“…According to whether a classifier is used to evaluate the goodness of candidate features in the feature selection process, feature selection techniques are typically divided into three categories: filter methods, wrapper methods and embedded methods [12]. Filter methods evaluate the quality of a feature using the intrinsic properties of training samples, so they have a lower computational complexity and better generalization ability and are flexible in combination with various classifiers [13,14]. In contrast to filter methods, wrapper methods are tightly coupled with a classifier to evaluate the quality of a candidate feature and often use the classification error rate as the evaluation criterion [12].…”