基于统计相关性与K-means的区分基因子集选择算法

针对高维小样本癌症基因数据集的有效区分基因子集选择难题,提出基于统计相关性和K-means的新颖混合基因选择算法实现有效区分基因子集选择.算法首先采用Pearson相关系数和Wilcoxon秩和检验计算各基因与类标的相关性,根据统计相关性原则选取与类标相关性较大的若干基因构成预选择基因子集;然后,采用K-means算法将预选择基因子集中高度相关的基因聚集到同一类簇,训练SVM分类模型,计算每一个基因的权重,从每一类簇选择一个权重最大或者采用轮盘赌思想从每一类簇选择一个得票数最多的基因作为本类簇的代表基因,各类簇的代表基因构成有效区分基因子集.将该算法与采用随机策略选择各类簇代表基因的随机基因选择算法Random, Guyon的经典基因选择算法SVM-RFE、采用顺序前向搜索策略的基因选择算法SVM-SFS进行实验比较,几个经典基因数据集上的200次重复实验的平均实验结果表明:所提出的混合基因选择算法能够选择到区分性能非常好的基因子集,建立在该区分基因子集上的分类器具有非常好的分类性能.     

Gene selection using genetic algorithm and support vectors machines

In this paper, we present a gene selection method based on genetic algorithm (GA) and support vector machines (SVM) for cancer classification. First, the Wilcoxon rank sum test is used to filter noisy and redundant genes in high dimensional microarray data. Then, the different highly informative genes subsets are selected by GA/SVM using different training sets. The final subset, consisting of highly discriminating genes, is obtained by analyzing the frequency of appearance of each gene in the different gene subsets. The proposed method is tested on three open datasets: leukemia, breast cancer, and colon cancer data. The results show that the proposed method has excellent selection and classification performance, especially for breast cancer data, which can yield 100% classification accuracy using only four genes.     

Identification of cancerous gene groups from microarray data by employing adaptive genetic and support vector machine technique

Nowadays, microarray gene expression data plays a vital role in tumor classification. However, due to the accessibility of a limited number of tissues compared to large number of genes in genomic data, various existing methods have failed to identify a small subset of discriminative genes. To overcome this limitation, in this paper, we developed a new hybrid technique for gene selection, called ensemble multipopulation adaptive genetic algorithm (EMPAGA) that can overlook the irrelevant genes and classify cancer accurately. The proposed hybrid gene selection algorithm comprises of two phase. In the first phase, an ensemble gene selection (EGS) method used to filter the noisy and redundant genes in high-dimensional datasets by combining multilayer and F-score approaches. Then, an adaptive genetic algorithm based on multipopulation strategy with support vector machine and naïve Bayes (NB) classifiers as a fitness function is applied for gene selection to select the extremely sensible genes from the reduced datasets. The performance of the proposed method is estimated on 10 microarray datasets of numerous tumor. The comprehensive results and various comparisons disclose that EGS has a remarkable impact on the efficacy of the adaptive genetic algorithm with multipopulation strategy and enhance the capability of the proposed approach in terms of convergence rate and solution quality. The experiments results demonstrate the superiority of the proposed method when compared to other standard wrappers regarding classification accuracy and optimal number of genes.     

Parallel multi-swarm optimizer for gene selection in DNA microarrays

The execution of many computational steps per time unit typical of parallel computers offers an important benefit in reducing the computing time in real world applications. In this work, a parallel Particle Swarm Optimization (PSO) is used for gene selection of high dimensional Microarray datasets. The proposed algorithm, called PMSO, consists of running a set of independent PSOs following an island model, where a migration policy exchanges solutions with a certain frequency. A feature selection mechanism is embedded in each subalgorithm for finding small samples of informative genes amongst thousands of them. PMSO has been experimentally assessed with different population structures on four well-known cancer datasets. The contributions are twofold: our parallel approach is able to improve sequential algorithms in terms of computational time/effort (Efficiency of 85%), as well as in terms of accuracy rate, identifying specific genes that our work suggests as significant ones for an accurate classification. Additional comparisons with several recent state the of art methods also show competitive results with improvements of over 100% in the classification rate and very few genes per subset.     

基于相容关系的基因选择方法

有效的基因选择是对基因表达数据进行分析的重要内容。粗糙集作为一种软计算方法能够保持在数据集分类能力不变的基础上,对属性进行约简。由于基因表达数据的连续性,为了避免运用粗糙集方法所必需的离散化过程带来的信息丢失,将相容粗糙集应用于基因的特征选取,提出了基于相容关系的基因选择方法。首先,通过i检验对基因表达数据进行排列,选择评分靠前的若干基因;然后,通过相容粗糙集对这些基因进一步约简。在两个标准的基因表达数据上进行了实验,结果表明该方法是可行性和有效性的。     

Optimal gene subset selection using the modified SFFS algorithm for tumor classification

A reliable and precise classification of tumors is essential for successful treatment of cancer. Gene selection is an important step for improved diagnostics. The modified SFFS (sequential forward floating selection) algorithm based on weighted Mahalanobis distance, called MSWM, is proposed to identify optimal informative gene subsets taking into account joint discriminatory power for accurate discrimination in this study. Firstly, we make use of the one-dimensional weighted Mahalanobis distance to perform a preliminary selection of genes and then make use of the modified SFFS method and multidimensional weighted Mahalanobis distance to obtain the optimal informative gene subset for tumor classification. Finally, we used the k nearest neighbor and naive Bayes methods to classify tumors based on the optimal gene subset selected using the MSWM method. To validate the efficiency, the proposed MSWM method is applied to classify two different DNA microarray datasets. Our empirical study shows that the MSWM method for tumor classification can obtain better effectiveness of classification than the BWR (the ratio of between-groups to within-groups sum of squares) and IVGA_I (independent variable group analysis I) methods. It suggests that the MSWM gene selection method is ability to obtain correct informative gene subsets taking into account genes’ joint discriminatory power for tumor classification.     

Augmentation of a nearest neighbour clustering algorithm with a partial supervision strategy for biomedical data classification

Abstract: In this paper, a partial supervision strategy for a recently developed clustering algorithm, the nearest neighbour clustering algorithm (NNCA), is proposed. The proposed method (NNCA-PS) offers classification capability with a smaller amount of a priori knowledge, where a small number of data objects from the entire data set are used as labelled objects to guide the clustering process towards a better search space. Experimental results show that NNCA-PS gives promising results of 89% sensitivity at 95% specificity when used to segment retinal blood vessels, and a maximum classification accuracy of 99.5% with 97.2% average accuracy when applied to a breast cancer data set. Comparisons with other methods indicate the robustness of the proposed method in classification. Additionally, experiments on parallel environments indicate the suitability and scalability of NNCA-PS in handling larger data sets.     

一种基于微阵列数据的集成分类方法*

针对现有的微阵列数据集成分类方法分类精度不高这一问题,提出了一种Bagging-PCA-SVM方法。该方法首先采用Bootstrap技术对训练样本集重复取样,构成大量训练样本子集,然后在每个子集上进行特征选择和主成分分析以消除噪声基因与冗余基因;最后利用支持向量机作为分类器,采用多数投票的方法预测样本的类属。通过三个数据集进行了测试,测试结果表明了该方法的有效性和可行性。     

A feature selection method using improved regularized linear discriminant analysis

Investigation of genes, using data analysis and computer-based methods, has gained widespread attention in solving human cancer classification problem. DNA microarray gene expression datasets are readily utilized for this purpose. In this paper, we propose a feature selection method using improved regularized linear discriminant analysis technique to select important genes, crucial for human cancer classification problem. The experiment is conducted on several DNA microarray gene expression datasets and promising results are obtained when compared with several other existing feature selection methods.     

On learning dual classifiers for better data classification

Instance selection aims at filtering out noisy data (or outliers) from a given training set, which not only reduces the need for storage space, but can also ensure that the classifier trained by the reduced set provides similar or better performance than the baseline classifier trained by the original set. However, since there are numerous instance selection algorithms, there is no concrete winner that is the best for various problem domain datasets. In other words, the instance selection performance is algorithm and dataset dependent. One main reason for this is because it is very hard to define what the outliers are over different datasets. It should be noted that, using a specific instance selection algorithm, over-selection may occur by filtering out too many ‘good’ data samples, which leads to the classifier providing worse performance than the baseline. In this paper, we introduce a dual classification (DuC) approach, which aims to deal with the potential drawback of over-selection. Specifically, performing instance selection over a given training set, two classifiers are trained using both a ‘good’ and ‘noisy’ sets respectively identified by the instance selection algorithm. Then, a test sample is used to compare the similarities between the data in the good and noisy sets. This comparison guides the input of the test sample to one of the two classifiers. The experiments are conducted using 50 small scale and 4 large scale datasets and the results demonstrate the superior performance of the proposed DuC approach over the baseline instance selection approach.     

一种基于特征选择的不平衡数据分类算法

不平衡数据分类是当前机器学习的研究热点,传统分类算法通常基于数据集平衡状态的前提,不能直接应用于不平衡数据的分类学习.针对不平衡数据分类问题,文章提出一种基于特征选择的改进不平衡分类提升算法,从数据集的不同类型属性来权衡对少数类样本的重要性,筛选出对有效预测分类出少数类样本更意义的属性,同时也起到了约减数据维度的目的.然后结合不平衡分类算法使数据达到平衡状态,最后针对原始算法错分样本权值增长过快问题提出新的改进方案,有效抑制权值的增长速度.实验结果表明,该算法能有效提高不平衡数据的分类性能,尤其是少数类的分类性能.     

基于MDA-RS算法的特征基因选取方法*

建立病变组织分类模型的关键在于找出一组能准确区分样本类别的特征基因。糙集理论中的属性依赖度分析方法能对目标数据进行有效分析。基于属性间的依赖关系和属性对决策的影响存在这样的关系,即属性依赖度越大,属性就越重要,对决策划分的影响就越大,提出了一种属性最大依赖度(maximum dependency of attributes based on rough sets,MDA-RS)算法,并将其应用于特征基因选取。首先用启发式K-均值聚类算法对基因进行聚类分析得到类数为k的基因子集;然后用MDA-RS选出每类的     

Classification consistency analysis for bootstrapping gene selection

Consistency modelling for gene selection is a new topic emerging from recent cancer bioinformatics research. The result of operations such as classification, clustering, or gene selection on a training set is often found to be very different from the same operations on a testing set, presenting a serious consistency problem. In practice, the inconsistency of microarray datasets prevents many typical gene selection methods working properly for cancer diagnosis and prognosis. In an attempt to deal with this problem, this paper proposes a new concept of classification consistency and applies it for microarray gene selection problem using a bootstrapping approach, with encouraging results.     

Prostate cancer biomarker discovery using high performance mass spectral serum profiling

Prostate-specific antigen (PSA) is the most widely used serum biomarker for early detection of prostate cancer (PCA). Nevertheless, PSA level can be falsely elevated due to prostatic enlargement, inflammation or infection, which limits the PSA test specificity. The objective of this study is to use a machine learning approach for the analysis of mass spectrometry data to discover more reliable biomarkers that distinguish PCA from benign specimens. Serum samples from 179 prostate cancer patients and 74 benign patients were analyzed. These samples were processed using ProXPRESSION™ Biomarker Enrichment Kits (PerkinElmer). Mass spectra were acquired using a prOTOF™ 2000 matrix-assisted laser desorption/ionization orthogonal time-of-flight (MALDI-O-TOF) mass spectrometer. In this study, we search for potential biomarkers using our feature selection method, the Extended Markov Blanket (EMB). From the new marker selection algorithm, a panel of 26 peaks achieved an accuracy of 80.7%, a sensitivity of 83.5%, a specificity of 74.4%, a positive predictive value (PPV) of 87.9%, and a negative predictive value (NPV) of 68.2%. On the other hand, when PSA alone was used (with a cutoff of 4.0 ng/ml), a sensitivity of 66.7%, a specificity of 53.6%, a PPV of 73.5%, and a NPV of 45.4% were obtained.     

Pattern classification in DNA microarray data of multiple tumor types

In this paper, we propose a genetic algorithm with silhouette statistics as discriminant function (GASS) for gene selection and pattern recognition. The proposed method evaluates gene expression patterns for discriminating heterogeneous cancers. Distance metrics and classification rules have also been analyzed to design a GASS with high classification accuracy. Moreover, the proposed method is compared to previously published methods. Various experimental results show that our method is effective for classifying the NCI60, the GCM and the SRBCTs datasets. Moreover, GASS outperforms other existing methods in both the leave-one-out cross validations and the independent test for novel data.     

A deep stacked wavelet auto-encoders to supervised feature extraction to pattern classification

The major issue in pattern classification is in the extraction of features in the training phase. The focus of this work is on combining the ability of wavelet networks and the deep learning techniques to propose a new supervised feature extraction method to pattern classification. This new approach allows the classification of all classes of the dataset by the reconstruction of a Deep Stacked wavelet Auto-Encoder. This Network is obtained after a series of wavelet Auto-Encoders followed by a Softmax classifier at the last layer. Finally, a fine-tuning is applied for the improvement of our result using a back propagation algorithm. Our approach is tested with different image datasets which are the COIL-100, the APTI and the ImageNet datasets and is also tested with two other audio corpuses that contain Arabic words and French words. The experimental test demonstrates the efficiency of our network for image and audio classification compared to other methods.     

Direct integration of microarrays for selecting informative genes and phenotype classification

The ability to provide thousands of gene expression values simultaneously makes microarray data very useful for phenotype classification. A major constraint in phenotype classification is that the number of genes greatly exceeds the number of samples. We overcame this constraint in two ways; we increased the number of samples by integrating independently generated microarrays that had been designed with the same biological objectives, and reduced the number of genes involved in the classification by selecting a small set of informative genes. We were able to maximally use the abundant microarray data that is being stockpiled by thousands of different research groups while improving classification accuracy. Our goal is to implement a feature (gene) selection method that can be applicable to integrated microarrays as well as to build a highly accurate classifier that permits straightforward biological interpretation. In this paper, we propose a two-stage approach. Firstly, we performed a direct integration of individual microarrays by transforming an expression value into a rank value within a sample and identified informative genes by calculating the number of swaps to reach a perfectly split sequence. Secondly, we built a classifier which is a parameter-free ensemble method using only the pre-selected informative genes. By using our classifier that was derived from large, integrated microarray sample datasets, we achieved high accuracy, sensitivity, and specificity in the classification of an independent test dataset.     

迭代再权q范数正则化LS SVM分类算法

提出一种迭代再权q范数正则化最小二乘支持向量机(LS SVM)分类算法。该算法通过交叉校验过程选择正则化范数的阶次q (0相似文献   

Kernel-based learning and feature selection analysis for cancer diagnosis

DNA microarray is a very active area of research in the molecular diagnosis of cancer. Microarray data are composed of many thousands of features and from tens to hundreds of instances, which make the analysis and diagnosis of cancer very complex. In this case, gene/feature selection becomes an elemental and essential task in data classification. In this paper, we propose a complete cancer diagnostic process through kernel-based learning and feature selection. First, support vector machines recursive feature elimination (SVM-RFE) is used to prefilter the genes. Second, the SVM-RFE is enhanced by using binary dragonfly (BDF), which is a recently developed metaheuristic that has never been benchmarked in the context of feature selection. The objective function is the average of classification accuracy rate generated by three kernel-based learning methods. We conducted a series of experiments on six microarray datasets often used in the literature. Experiment results demonstrate that this approach is efficient and provides a higher classification accuracy rate using a reduced number of genes.     

An expert system to classify microarray gene expression data using gene selection by decision tree

Gene selection can help the analysis of microarray gene expression data. However, it is very difficult to obtain a satisfactory classification result by machine learning techniques because of both the curse-of-dimensionality problem and the over-fitting problem. That is, the dimensions of the features are too large but the samples are too few. In this study, we designed an approach that attempts to avoid these two problems and then used it to select a small set of significant biomarker genes for diagnosis. Finally, we attempted to use these markers for the classification of cancer. This approach was tested the approach on a number of microarray datasets in order to demonstrate that it performs well and is both useful and reliable.