A Review of Feature Extraction Software for Microarray Gene Expression Data

Table 1

A summary for PCA software.

Number	Software	Author/year	Language	Features
1	FactoMineR	Lê et al. [5]	R	(i) Various dimension reduction methods such as PCA, CA, and MCA; (ii) Different types of variables, data structures, and supplementary information are considered; (iii) The PCA function can handle missing values
2	ExPosition	Beaton et al. [8]	R	(i) Numerous multivariate analysis methods such as PCA and Generalized Principal Component Analysis (GPCA); (ii) Can be applied to quantitative and qualitative data; (iii) Implementation of Singular Value Decomposition
3	amap	Lucas [9]	R	(i) Different types of PCA are provided: PCA, Generalized PCA, and Robust PCA; (ii) Clustering methods are provided such as hierarchical clustering and -means clustering; (iii) Plotting function for PCA; (iv) Computing distance and dissimilarity matrices
4	ADE-4	Thioulouse et al. [10]	R	A variety of methods such as PCA, CA, Principal Analysis Regression, PLS, and others are offered
5	MADE4	Culhane et al. [11]	R	(i) Functions provided by ADE-4; (ii) Integration of multiple datasets for multivariate analysis; (iii) Functions for visualizing and plotting the results of analysis, including 3D plots; (iv) Addition of LLSimpute algorithm for imputation of missing values
6	XLMiner	Witten and Frank [12]	Implemented in Excel	(i) Provision of data reduction methods such as PCA; (ii) Can be used for classification, clustering, data preprocessing, data normalization, and others
7	ViSta	Young et al. [13]	C++, Fortran, XLisp, and ViDAL	(i) Multivariate analysis methods are offered such as PCA, Interactive Cluster Analysis, and Parallel Boxplots; (ii) Provision of dynamic and high-interaction visualization for displaying multiple views of data
8	imDEV	Grapov and Newman [14]	Visual Basic and R	(i) Data preprocessing: missing values imputation and data transformations; (ii) Clustering methods are offered; (iii) Dimension reduction methods: PCA and ICA; (iv) Feature selection methods; (v) Visualization of data dependencies
9	Statistics Toolbox	The MathWorks [15]	MATLAB	(i) Multivariate statistics such as PCA, clustering, and others; (ii) Statistical plots, probability distributions, linear models, nonlinear models for regression, and others are provided
10	Weka	Hall et al. [16]	Java	A variety of machine learning algorithms are provided such as feature selection, data preprocessing, regression, dimension reduction, classification, and clustering methods
11	NAG Library	NAG Toolbox for MATLAB; [17]	Fortran and C	(i) Provision of more than 1700 mathematical and statistical algorithms; (ii) Multivariate analysis using PCA can be implemented using the g03aa routine