Fig 6 - uploaded by Ebenezer Olusegun George
Content may be subject to copyright.
showsSSEwith100multiplerunswithindependent initial starting values with Gaussian distribution. Fig. 7 provides the Histogram of the SSE for 100 runs for the testingdata.Bothfiguresshowthatthestandarddeviationis
Source publication
We propose Bayesian neural networks (BNN) with structural learning
for exploring microarray data in gene expressions. The approach employs
representative data and regularization to capture correlation among gene
expressions and Bayesian techniques to extract gene expression
information from noisy data. The performance was verified with
stratified c...
Similar publications
Preoperative chemoradiotherapy (CRT) has become a widely used treatment for improving local control of disease and increasing survival rates of rectal cancer patients. We aimed to identify a set of genes that can be used to predict responses to CRT in patients with rectal cancer.
Gene expression profiles of pre-therapeutic biopsy specimens obtained...
Citations
... In recent years, some microarray image gridding methods presented [4] [5] [10] [11] [12] had more accuracy but more computation complexity than the precedent ones. In this paper, we presented one serialized microarray image gridding method based on image projection sequences power spectral estimations and local maxima searching for projection sequences, which promoted gridding accuracy and decreased computation complexity significantly. ...
... Over recent years, different classification approaches have been applied to micro array gene expression data classification [5], [11], [12], [13], [14], [15], [16]. Previous work often shows that neural networks and SVMs outperformed the other approaches used [2], [3], [4], [6], [7], [8], [9]. ...
Classification of yeast genes based on their expression levels obtained from micro array hybridization experiments is an important and challenging application domain in data mining and knowledge discovery. Over the past decade, neural networks and support vector machines (SVMs) have achieved good results for genes classification. This paper presents a methodology which uses two neural networks to classify unseen genes based on their expression levels. In order to remove some of the noise and deal with the imbalanced class distribution of the dataset, data pre-processing is firstly performed before data classification in which data cleaning, data transformation and data over-sampling using SMOTE algorithm are undertaken. Thereafter, two neural networks with different architectures are trained using Scaled Conjugate Gradient in two different ways: 1) the training-validation-testing approach and 2) 10-fold cross-validation. Experimental results show that this methodology outperforms the previous best-performing SVM for this problem and 8 other classifiers: 3 SVMs, C4.5, Bayesian network, Naive Bayes, K-NN and JRip.
... Comparison of discrimination methods for the classification of tumors using gene expression data was done by Dudoit et al. (2002). We reported a Bayesian neural network approach earlier (Liang et al., 2002). Previous study showed that traditional statistical models can provide some insight into gene expressions and has precise results, but the weaknesses of statistical models are that they can not capture the dynamic changing of gene expressions from time to time well and are sensitive to noise and assumptions. ...
Heterogeneous gene expressions provide insight into the biological role of gene interaction with the environment, disease development and drug effect at the molecular level. We propose Time Lagged Recurrent Neural Network with trajectory learning for identifying and classifying gene functional patterns from the heterogeneous nonlinear time series microarray experiments. The proposed procedures identify gene functional patterns from the dynamics of a state-trajectory learned in the heterogeneous time series and the gradient information over time. Trajectory learning with Back-propagation through time algorithm can recognise gene expression patterns vary over time. This reveals more information about the regulatory network underlying gene expressions.
... The Bayesian neural network is used with structural learning with forgetting for searching optimal network size and structure of microarray data in order to capture the structural information of gene expressions [43]. The process of Bayesian learning starts with a feed forward neural network (FFNN) and prior distribution for the network parameters. ...
Different bioinformatics tasks like gene sequence analysis, gene finding, protein structure prediction and analysis, gene expression with microarray analysis and gene regulatory network analysis are described along with some classical approaches. The relevance of intelligent systems and neural networks to these problems is mentioned. Different neural network based algorithms to address the aforesaid tasks are then presented. Finally some limitations of the current research activity are provided. An extensive bibliography is included.
... This can be done in two ways. Firstly, the network can be regularized by setting the network parameters to some prior distribution (see Liang et al., 2002). Secondly, the Bayesian approach can be used in developing the training algorithms for the MLP. ...
The aim of this thesis was to evaluate the usability of self-organizing maps and some other methods of computational intelligence in analysing and modelling problems of environmental informatics and bioinformatics. The concepts of environmental informatics, bioinformatics, computational intelligence and data mining are first defined. There follows an introduction to the data processing chain of knowledge discovery and the methods used in this thesis, namely linear regression, self-organizing maps (SOM), Sammon's mapping, U-matrix representation, fuzzy logic, c-means and fuzzy c-means clustering, multi-layer perceptron (MLP), and regularization and Bayesian techniques. The challenges posed by environmental processes and bioprocesses are then identified, including missing data problems, complex lagged dependencies among variables, non-linear chaotic dynamics, ill-defined inverse problems, and large search space in optimization tasks. The works included in this thesis are then evaluated and discussed. The results show that the combination of SOM and Sammon's mapping has great potential in data exploration, and can be used to reveal important features of the measurement techniques (e.g. separability of compounds), reveal new information about already studied phenomena, speed up research work, act as a hypothesis generator for traditional research, and supply clear and intuitive visualization of the environmental phenomenon studied. The results of regression studies show, as expected, that the MLP network yields better estimates in predicting future values of airborne pollutant concentration of NO2 compared with SOM based regression or the least squares approach using periodic components. Additionally, the use of local MLP models is shown to be slightly better for estimating future values of episodes compared with one MLP model only. However, it can be concluded in general that the architectural issues tested are not able to solve solely model performance problems. Finally, recommendations for future work are laid out. Firstly, the data exploration solution should be enhanced with methods from signal processing to enable the handling of measurements with different time scale and lagged multivariate time-series. The main suggestion, however, is to create an integrated environment for testing different hybrid schemes of computational intelligence for better time-series forecasting in environmental informatics and bioinformatics. Kuopio University publications. C, Natural and environmental sciences, ISSN 1235-0486; 167
... Unsupervised neural network and associated memory neural network was done by Azuaje [2] and Bicciato, Pandin, Didone, and Bello [3]. We reported a Bayesian neural network approach earlier [17]. Previous study showed that traditional statistical models can provide some insight into gene expressions and has precise results, but the weaknesses of statistical models are that they can not capture the dynamic changing of gene expressions from time to time well and are sensitive to noise and assumptions. ...
Heterogeneous types of gene expressions may provide a better insight into the biological role of gene interaction with the environment, disease development and drug effect at the molecular level. In this paper for both exploring and prediction purposes a Time Lagged Recurrent Neural Network with trajectory learning is proposed for identifying and classifying the gene functional patterns from the heterogeneous nonlinear time series microarray experiments. The proposed procedures identify gene functional patterns from the dynamics of a state-trajectory learned in the heterogeneous time series and the gradient information over time. Also, the trajectory learning with Backpropagation through time algorithm can recognize gene expression patterns vary over time. This may reveal much more information about the regulatory network underlying gene expressions. The analyzed data were extracted from spotted DNA microarrays in the budding yeast expression measurements, produced by Eisen et al. The gene matrix contained 79 experiments over a variety of heterogeneous experiment conditions. The number of recognized gene patterns in our study ranged from two to ten and were divided into three cases. Optimal network architectures with different memory structures were selected based on Akaike and Bayesian information statistical criteria using two-way factorial design. The optimal model performance was compared to other popular gene classification algorithms such as Nearest Neighbor, Support Vector Machine, and Self-Organized Map. The reliability of the performance was verified with multiple iterated runs.
Microarrays are revolutionary high-throughput tools for the study of gene and proteomics expressions. This technology has been rapidly advanced in recent years. Image-based analysis is an important aspect of microarray experiments, which can have a potentially large impact on subsequent analyses and therefore becomes the research hot spot. Numbers of analysis methods used on microarray data have been issued, including gridding, segmentation, data extraction and expression, image correction, noise removing and so on. In this paper, all classes of image based analysis methods are presented and summarized. In addition, we compare and analyze their performance and results with a lot of practical instances.
Microarray image gridding is one important step of microarray image analysis to determine 2D image coordinates of all array spots in the hybridized gene chip image. Accuracy of microarray image gridding will affect the reliability of gene-chip data extraction and even the final analysis results of gene-chip assays. However, in recent years, many new microarray image gridding methods presented have higher accuracy but have more computation complexity than those precedented ones. To promote gridding accuracy and decrease computation complexity simultaneously, we presented a novel and simple microarray image gridding method based on image projection sequences analysis and local extrema searching. Firstly we transformed 2D microarray image into vertical and horizontal 1D projection sequences, secondly utilized signal processing methods of low pass filtering and zero mean to filtered projection sequences, thirdly computed the first-order difference and second-order difference for the smoothed signals, and finally realized microarray image gridding according to the local extrema of difference sequences and span information of spots array on the microarray image. The subsequent gridding experiments showed that this method had met the requirements of computing accuracy and efficiency of microarray image gridding.
Based on the theory of infrared imaging, the infrared detection technology and the theory of thermal conductivity inversion, we developed the Mine Data Acquisition systems of the IR Thermometers. By the technology, detecting the temperature of infrared thermal imaging zone of in the gobs of Dongtan Coal Mine, we confirmed the temperature, and then judged the conditions of coal spontaneous combustion to achieve the early forecast of coal spontaneous combustion. And through detecting the temperature of the infrared thermal imaging zone in the track crossheading 1303, the gob (extracting stage), along the empty side of the track crossheading1303 and stopping coal mining face, to ensure the normal safety in production on working face, playing the features of the advanced hazard identification technology of infrared thermal imaging.
