Cheng-San Yang

Chemical Composition and Antibacterial Activities of Illicium verum Against Antibiotic-Resistant Pathogens

In recent years, human pathogenic microorganisms have developed multiple drug resistance and caused serious nosocomial infections. In this study, we identified four new antimicrobial compounds from the Chinese herbal medicine Illicium verum and assessed their antibacterial efficacies. The supercritical CO₂ and ethanol extracts of Illicium verum showed substantial antibacterial activity against 67 clinical drug-resistant isolates, including 27 Acinetobacter baumannii, 20 Pseudomonas aeruginosa, and 20 methicillin-resistant Staphylococcus aureus. The diethyl ether (EE) fraction obtained from partition extraction and supercritical CO₂ extracts revealed an antibacterial activity with a minimum inhibitory concentration value of 0.15-0.70 mg/mL and 0.11 mg/mL, respectively. The EE fraction of I. verum showed synergetic effects with some commercial antibiotics. The antimicrobial mechanism was investigated with killing curves and scanning electron microscopy observation. The chemical components of the extracts were analyzed by spectrophotometry; (E)-anethole, anisyl acetone, anisyl alcohol, and anisyl aldehyde exhibited antibacterial activity against different clinical isolates. These extracts from I. verum can be further developed into antibiotic medicines due to their proven antibacterial activity.

Gene selection and classification using Taguchi chaotic binary particle swarm optimization

The purpose of gene expression analysis is to discriminate between classes of samples, and to predict the relative importance of each gene for sample classification. Microarray data with reference to gene expression profiles have provided some valuable results related to a variety of problems and contributed to advances in clinical medicine. Microarray data characteristically have a high dimension and a small sample size. This makes it difficult for a general classification method to obtain correct data for classification. However, not every gene is potentially relevant for distinguishing the sample class. Thus, in order to analyze gene expression profiles correctly, feature (gene) selection is crucial for the classification process, and an effective gene extraction method is necessary for eliminating irrelevant genes and decreasing the classification error rate. In this paper, correlation-based feature selection (CFS) and the Taguchi chaotic binary particle swarm optimization (TCBPSO) were combined into a hybrid method. The K-nearest neighbor (K-NN) with leave-one-out cross-validation (LOOCV) method served as a classifier for ten gene expression profiles. Experimental results show that this hybrid method effectively simplifies features selection by reducing the number of features needed. The classification error rate obtained by the proposed method had the lowest classification error rate for all of the ten gene expression data set problems tested. For six of the gene expression profile data sets a classification error rate of zero could be reached. The introduced method outperformed five other methods from the literature in terms of classification error rate. It could thus constitute a valuable tool for gene expression analysis in future studies.

Boolean binary particle swarm optimization for feature selection

Feature selection is the process of choosing a subset of features from an original set. This subset should be necessary, reasonably represent the original data, and useful for identification classification. The task of feature selection is to search for an optimal solution in a - usually large - search space. However, if the search space too large, difficulties can occur during the search process, often resulting in a considerable increase in computational time. A particle swarm optimization algorithm (PSO) is a relatively new evolutionary computation technique, which has previously been used to implement feature selection. However, particle swarm optimization, like other evolutionary algorithms, tends to converge at a local optimum early. In this paper, we introduce a Boolean function which improves on the disadvantages of standard particle swarm optimization and use it to implement a feature selection for six microarray data sets. The experimental results show that the proposed method selects a smaller number of feature subsets and obtains better classification accuracy than standard PSO.

Chaotic maps in binary particle swarm optimization for feature selection

Feature selection is a useful pre-processing technique for solving classification problems. The challenge of using evolutionary algorithms lies in solving the feature selection problem caused by the number of features. Classification data may contain useless, redundant or misleading features. To increase the classification accuracy, the primary objective is to remove irrelevant features in the feature space and identify the relevant features. Binary particle swarm optimization (BPSO) has been applied successfully in solving feature selection problem. In this paper, two kinds of chaotic maps are embedded in binary particle swarm optimization (BPSO), a logistic map and a tent map, respectively. The purpose of the chaotic maps is to determine the inertia weight of the BPSO. In this study, we propose the chaotic binary particle swarm optimization (CBPSO) method to implement feature selection, and the K-nearest neighbor (K-NN) method with leave-one-out cross-validation (LOOCV) serves as a classifier to evaluate the classification accuracies. The proposed method showed promising results for feature selection with respect to the number of feature subsets. The classification accuracy obtained by the proposed method is superior to ones obtained by the other methods from the literature.

Feature Selection Using Memetic Algorithms

The feature selection process can be considered a problem of global combinatorial optimization in machine learning, which reduces the number of features, removes irrelevant, noisy and redundant data, and results in acceptable classification accuracy. In this study, we propose a combined filter method (ReliefF) and a wrapper method (memetic algorithm, MA) for classification. The goal of our method is to filter the irrelevant features and select the most important feature subsets. We used the ReliefF algorithm to calculate and update the scores of every feature for each data set, and then applied a MA for feature selection. The K-nearest neighbor (K-NN) method with leave-one-out cross-validation (LOOCV) serves as a classifier for evaluating classification accuracies. The experimental results show that the proposed method is superior to existing methods in terms of classification accuracy.

Genotype and Antibiotic Susceptibility Patterns of Drug-Resistant Pseudomonas aeruginosa and Acinetobacter baumannii Isolates in Taiwan

Rapid and accurate identification of the drug susceptibility profile of clinical strains is very important for controlling bacterial infections and determining the antibiotic therapy. The objective of this study was to investigate the spectrum of the correlation between phenotypic and genetic characters of the drug-resistant clinical isolates. A total of 133 clinical isolates, including 76 Acinetobacter baumannii and 57 Pseudomonas aeruginosa, were examined for their antibiotic susceptibility by the method of disc diffusion. Among them, most of the isolates were multiresistant, and 80% of the strains showed phenotypic resistance to beta-lactam antibiotics. Using PCR analysis, among the several types of beta-lactamases, TEM was the most prevalent, and OXA was the second most prevalent. The integron harbored was identified by conserved segment PCR, and 50% of the test isolates carried integrons with various gene cassette sizes inserted. The results obtained from this study reveal that the majority of these isolates displayed multiple drug resistance phenotypes that were associated with their mutational gene profiles.

An efficiency analysis of high-order combinations of gene–gene interactions using multifactor-dimensionality reduction

BackgroundMultifactor dimensionality reduction (MDR) is widely used to analyze interactions of genes to determine the complex relationship between diseases and polymorphisms in humans. However, the astronomical number of high-order combinations makes MDR a highly time-consuming process which can be difficult to implement for multiple tests to identify more complex interactions between genes. This study proposes a new framework, named fast MDR (FMDR), which is a greedy search strategy based on the joint effect property.ResultsSix models with different minor allele frequencies (MAFs) and different sample sizes were used to generate the six simulation data sets. A real data set was obtained from the mitochondrial D-loop of chronic dialysis patients. Comparison of results from the simulation data and real data sets showed that FMDR identified significant gene–gene interaction with less computational complexity than the MDR in high-order interaction analysis.ConclusionFMDR improves the MDR difficulties associated with the computational loading of high-order SNPs and can be used to evaluate the relative effects of each individual SNP on disease susceptibility. FMDR is freely available at http://bioinfo.kmu.edu.tw/FMDR.rar.

Chaotic Genetic Algorithm for Gene Selection and Classification Problems

Pattern recognition techniques suffer from a well-known curse, the dimensionality problem. The microarray data classification problem is a classical complex pattern recognition problem. Selecting relevant genes from microarray data poses a formidable challenge to researchers due to the high-dimensionality of features, multiclass categories being involved, and the usually small sample size. The goal of feature (gene) selection is to select those subsets of differentially expressed genes that are potentially relevant for distinguishing the sample classes. In this paper, information gain and chaotic genetic algorithm are proposed for the selection of relevant genes, and a K-nearest neighbor with the leave-one-out crossvalidation method serves as a classifier. The chaotic genetic algorithm is modified by using the chaotic mutation operator to increase the population diversity. The enhanced population diversity expands the GAs search ability. The proposed approach is tested on 10 microarray data sets from the literature. The experimental results show that the proposed method not only effectively reduced the number of gene expression levels, but also achieved lower classification error rates than other methods.

Tag SNP selection using particle swarm optimization.

Single nucleotide polymorphisms (SNPs) are the most abundant form of genetic variations amongst species. With the genome‐wide SNP discovery, many genome‐wide association studies are likely to identify multiple genetic variants that are associated with complex diseases. However, genotyping all existing SNPs for a large number of samples is still challenging even though SNP arrays have been developed to facilitate the task. Therefore, it is essential to select only informative SNPs representing the original SNP distributions in the genome (tag SNP selection) for genome‐wide association studies. These SNPs are usually chosen from haplotypes and called haplotype tag SNPs (htSNPs). Accordingly, the scale and cost of genotyping are expected to be largely reduced. We introduce binary particle swarm optimization (BPSO) with local search capability to improve the prediction accuracy of STAMPA. The proposed method does not rely on block partitioning of the genomic region, and consistently identified tag SNPs with higher prediction accuracy than either STAMPA or SVM/STSA. We compared the prediction accuracy and time complexity of BPSO to STAMPA and an SVM‐based (SVM/STSA) method using publicly available data sets. For STAMPA and SVM/STSA, BPSO effective improved prediction accuracy for smaller and larger scale data sets. These results demonstrate that the BPSO method selects tag SNP with higher accuracy no matter the scale of data sets is used.

Comparative Particle Swarm Optimization (CPSO) for solving optimization problems

Particle swarm optimization (PSO) is a stochastic and population-based intelligence search algorithm, which has been demonstrated to solve optimization problem effectively. However, as the particle properties become increasingly similar after several generations, the particles tend to cluster around the best (fittest) particle in the swarm, which results in premature convergence of the PSO algorithm. In other words, the particles get trapped in the local optimal solution. In this paper, a new conception of PSO is proposed, which is based on comparing the experience of all particles in the swarm to generate a better position, and guide all particles toward the best possible solution. Experiments conducted on three benchmark functions show that the new algorithm is more efficient than standard PSO.