SlimPLS-package(SlimPLS)
SlimPLS-package()所属R语言包:SlimPLS
slimPLS - Partial Least Squares based multivariate feature selection.
slimPLS - 偏最小二乘法为基础的多元功能选择。
译者:生物统计家园网 机器人LoveR
描述----------Description----------
This package introduces a multivariate feature selection algorithm developed in the Computational Genomics lab of Tel-Aviv University primarily for dimension reduction of gene expression data for the purpose of classification. The package defines methods for reading an expression data from a file, and extracting features using SlimPLS from such a matrix. For the purposes of testing it also provides a wrapper to a number of classification packages, allowing for learning and predicting of samples' classes learned after the SlimPLS dimension reduction. For comparison, it also provides an interface for a number of univariate feature selection methods based on scoring.
这个包在特拉维夫大学的降维的基因表达数据分类的目的主要是计算基因组学实验室开发的,引入了多元的特征选择算法。包定义读取文件中的数据的表达和提取功能,使用SlimPLS从这样一个矩阵的方法。为了测试的目的,它也提供了一个包装到一些分类包,使学习和预测样本的类的经验教训后SlimPLS降维。为了便于比较,它也提供了一个接口,一个单变量的特征选择方法的基础上评分。
Details
详细信息----------Details----------
The usual flow of using SlimPLS starts at reading an expression matrix from a standard tab delimited file into an ExpMat object, by calling:
常见的流程的使用SlimPLS开始从一个标准的制表符分隔的文件中读取表达矩阵ExpMat对象,通过调用:
my_exp_mat <- readExpMat("myFile.txt")
my_exp_mat <- readExpMat("myFile.txt")
Next, a set of feature is selected from this expression matrix by calling either
接着,选择一组特征可以通过调用从这个表达式矩阵
my_feature_set <- selectFeatures(expMat,...)
my_feature_set <- selectFeatures(expMat,...)
or
或
my_feature_set <- selectFeaturesSlimPLS(expMat,...). Then the features and the matrix are used to learn a classification model using
my_feature_set <- selectFeaturesSlimPLS(expMat,...)。然后,功能和矩阵用于学习的分类的模型中使用
my_model <- trainClassifier(...).
my_model <- trainClassifier(...)。
Last, the model learned can be used to predict classes of another (or the same) expression matrix, using
最新了解到,模型可用于预测的另一个(或同一个)表达矩阵的类,用
getClassification(...).
getClassification(...)。
The method is used for binary classification (i.e. two classes only), and is supervised (i.e. one needs to supply a labeling of samples as belonging to one of two classes)
方法用于二元分类(即只有两个类),并接受监督(即需要提供一个标签的样品,属于两班)
(作者)----------Author(s)----------
The method was developed by Michael Gutkin, Prof. Ron Shamir and Prof. Gideon Dror, and programming was
done by Michael Gutkin. It is maintained by Ofer Lavi.
参考文献----------References----------
1. Gutkin, M., MSc thesis, April 2008, Feature selection methods for classification of gene expression profiles, Tel Aviv Univetsity, Tel Aviv.
2. Gutkin, M., Shamir, R., Dror, G., 2009 SlimPLS: A Method for Feature Selection in Gene Expression-Based Disease Classification. PLoS ONE 4(7): e6416. doi:10.1371/journal.pone.0006416
转载请注明:出自 生物统计家园网(http://www.biostatistic.net)。
注:
注1:为了方便大家学习,本文档为生物统计家园网机器人LoveR翻译而成,仅供个人R语言学习参考使用,生物统计家园保留版权。
注2:由于是机器人自动翻译,难免有不准确之处,使用时仔细对照中、英文内容进行反复理解,可以帮助R语言的学习。
注3:如遇到不准确之处,请在本贴的后面进行回帖,我们会逐渐进行修订。
|