Ito Wasito

Iterative local Gaussian clustering for expressed genes identification linked to malignancy of human colorectal carcinoma

Gene expression profiling plays an important role in the identification of biological and clinical properties of human solid tumors such as colorectal carcinoma. Profiling is required to reveal underlying molecular features for diagnostic and therapeutic purposes. A non-parametric density-estimation-based approach called iterative local Gaussian clustering (ILGC), was used to identify clusters of expressed genes. We used experimental data from a previous study by Muro and others consisting of 1,536 genes in 100 colorectal cancer and 11 normal tissues. In this dataset, the ILGC finds three clusters, two large and one small gene clusters, similar to their results which used Gaussian mixture clustering. The correlation of each cluster of genes and clinical properties of malignancy of human colorectal cancer was analysed for the existence of tumor or normal, the existence of distant metastasis and the existence of lymph node metastasis.

Robust consensus clustering for identification of expressed genes linked to malignancy of human colorectal carcinoma

Previous studies have been conducted in gene expression profiling to identify groups of genes that characterize the colorectal carcinoma disease. Despite the success of previous attempts to identify groups of genes in the progression of the colorectal carcinoma disease, their methods either require subjective interpretation of the number of clusters, or lack stability during different runs of the algorithms. All of which limits the usefulness of these methods. In this study, we propose an enhanced algorithm that provides stability and robustness in identifying differentially expressed genes in an expression profile analysis. Our proposed algorithm uses multiple clustering algorithms under the consensus clustering framework. The results of the experiment show that the robustness of our method provides a consistent structure of clusters, similar to the structure found in the previous study. Furthermore, our algorithm outperforms any single clustering algorithms in terms of the cluster quality score.

Intelligent K-Means clustering for expressed genes identification linked to malignancy of human colorectal carcinoma

Cancer is one kind of deadly disease. Colorectal carcinoma is one type of cancer which is difficult to detect in its early stage. It has dangerous malignancy in its advance stage. Identify gene expressed and cancer linked to phenotype is an effort to identify and analyze correlation of genes and clinical phenotype (metastasis). In this paper Intelligent K-Means is used to cluster genes expression. It is a non parametric clustering that more powerful and more stable than GMM clustering which is used in previous research. After getting clusters of genes, then correlation ratio is used to identify whether genes in a cluster has a correlation with clinical metastasis. As the result in this paper, genes in cluster C and cluster E have correlation with normal-cancer tissue metastasis and distant metastasis. But, there is no cluster of genes has correlation with lymph node metastasis.

KNN-kernel based clustering for spatio-temporal database

Extracting and analyzing the interesting patterns from spatio-temporal databases, have drawn a great interest in various fields of research. Recently, a number of experiments have explored the problem of spatial or temporal data mining, and some clustering algorithms have been proposed. However, not many studies have been dealing with the integration of spatial data mining and temporal data mining. Moreover, the data in spatial temporal database can be categorized as high-dimensional data. Current density-based clustering might have difficulties with complex data sets including high-dimensional data. This paper presents Iterative Local Gaussian Clustering (ILGC), an algorithm that combines K-nearest neighbour (KNN) density estimation and Kernel density estimation, to cluster the spatiotemporal data. In this approach, the KNN density estimation is extended and combined with Kernel function, where KNN contributes in determining the best local data iteratively for kernel density estimation. The local best is defined as the set of neighbour data that maximizes the kernel function. Bayesian rule is used to deal with the problem of selecting the best local data. This paper utilized Gaussian kernel which has been proven successful in the clustering. To validate the KNN-kernel based algorithm, we compare its performance againts other popular algorithms, such as Self Organizing Maps (SOM) and K-Means, on Crime database. Results show that KNN-kernel based clustering has outperformed others.

Kernel based integration of Gene expression and DNA copy number

Kernel based integration data of gene expression and DNA copy number is used to analyze pattern of genes in breast cancer cell line. The integration data is clustered without any information about the number of k clusters. This paper proposes the use of intelligent kernel K-Means that is developed by combining intelligent K-Means and kernel K-Means. The technique is used to cluster data integration of gene expression and DNA copy number. The experiment results show that there are three clusters are successes to be found. Evaluation measure produce R value is 0.29.

Identification of Gene Expression Linked to Malignancy of Human Colorectal Carcinoma using Restricted Boltzmann Machines

Learning hidden information or pattern on gene expression data to uncover an underlying molecular features is called gene expression profiling. To perform gene expression profiling, an unsupervised machine learning method can be employed. In this paper, Gaussian RBM is proposed to obtain the optimal number of clusters and their members on human colorectal cancer dataset provided by Muro. Gaussian RBM forms two large numbers of genes clusters and one smaller cluster which has several tumour-classifier genes as its members. The two large numbers of genes clusters formed by Gaussian RBM succeed in showing a significant correlation with the existence of tumour and distant metastasis but they show no significant correlation with lymph node metastasis existence. The smaller number of genes clusters gives a statistically significant result in clustering patients into two groups.

Cancer subtype identification using deep learning approach

In this paper, a framework using deep learning approach is proposed to identify two subtypes of human colorectal carcinoma cancer. The identification process uses information from gene expression and clinical data which is obtained from data integration process. One of deep learning architecture, multimodal Deep Boltzmann Machines (DBM) is used for data integration process. The joint representation gene expression and clinical is later used as Restricted Boltzmann Machines (RBM) input for cancer subtype identification. Kaplan Meier survival analysis is employed to evaluate the identification result. The curves on survival plot obtained from Kaplan Meier analysis are tested using three statistic tests to ensure that there is a significant difference between those curves. According to Log Rank, Generalized Wilcoxon and Tarone-Ware, the two groups of patients with different cancer subtypes identified using the proposed framework are significantly different.

Adaptive genetic algorithm for reliable training population in plant breeding genomic selection

Many algorithms are developed to model Genomic Estimated Breeding Value (GEBV). Modeling GEBV evolves a huge size of genotype in both terms of the dimension (columns) and the instances (rows). Good combinations of features help in predicting which phenotype is being represented. Preparing a good training population sample is assumed to be a convenient solution to deal with such complex genotype data. In this research, an Adaptive Genetic Algorithm (AGA) is proposed. The adaptive characteristic of AGA by adjusting probabilities in crossover and mutation is expected to converge into the global optimum without getting trapped in local optima. The proposed method using AGA to optimize the feature selection and shrinkage mechanism is looked forward to provide a reliable model to be reused in other similar datasets.

Fully unsupervised clustering in nonlinearly separable data using intelligent Kernel K-Means

Intelligent Kernel K-Means is a fully unsupervised clustering technique. This technique is developed by combining Intelligent K-Means and Kernel K-Means. Intelligent Kernel K-Means used to cluster kernel matrix without any information about the number of clusters. The goal of this research is to evaluate the performance of Intelligent Kernel K-Means for clustering nonlinearly separable data. Various artificial nonlinearly separable data are used in this experiment. The best result is the clustering often ring datasets. It produces Adjusted Rand Index (ARI) = 1.

Evaluation of data integration strategies based on kernel method of clinical and microarray data.

The cancer classification problem is one of the most challenging problems in bioinformatics. The data provided by Netherland Cancer Institute consists of 295 breast cancer patient; 101 patients are with distant metastases and 194 patients are without distant metastases. Combination of features sets based on kernel method to classify the patient who are with or without distant metastases will be investigated. The single data set will be compared with three data integration strategies and also weighted data integration strategies based on kernel method. Least Square Support Vector Machine (LS-SVM) is chosen as the classifier because it can handle very high dimensional features, for instance, microarray data. The experiment result shows that the performance of weighted late integration and the using of only microarray data are almost similar. The data integration strategy is not always better than using single data set in this case. The performance of classification absolutely depends on the features that are used to represent the object.