Shigeomi Hara

A clustering method of chromosome fluorescence profiles using modified self organizing map controlled by simulated annealing

The clustering method by the self organizing map algorithm of chromosome profiles measured by a slit-scan flow-cytometer is proposed. Moreover, the physical models of chromosomes have been introduced in order to take into account the rotation of chromosomes in the flow-cytometer. By this modification, the lengths of chromosomes and the intensity distribution of chromosome fluorescence can be estimated from chromosome profile data measured by the flow-cytometer. But the clustering results did not converge identically in some experiments and the distribution of the rotation angles was unnatural. So, we introduced simulated annealing to improve the convergence of our SOM algorithm. We compared the clustering results of this method with those of the K-means method and the SOM method.

An application of genetic algorithm to DNA sequencing by oligonucleotide hybridization

The authors propose a sequencing algorithm for oligonucleotide hybridization using the genetic algorithm. The target DNA sequence reconstructed by the hybridization method is relatively long for genetic algorithm (GA), so special setups of the genetic operation are necessary. The authors introduce the grouping GA and a special crossover method for this problem. They carried out some experiments of sequence reconstruction, and examined the reconstructed sequences by comparing the motif length subsequences.

Application of Supervised Pareto Learning Self Organizing Maps and Its Incremental Learning

We have proposed Supervised Pareto Learning Self Organizing Maps(SP-SOM) based on the concept of Pareto optimality for the integration of multiple vectors and applied SP-SOM to the biometric authentication system which uses multiple behavior characteristics as feature vectors. In this paper, we examine performance of SP-SOM for the generic classification problem using iris data set. Furthermore, we propose the incremental learning algorithm for SP-SOM and examine effectiveness in a classification problem and adaptation ability to the change of the behavior biometric features by time.

A design method of DNA chips for SNP analysis using self-organizing maps

We introduce a design method of DNA chips using self-organizing maps (SOM). DNA chips are powerful tools for sequencing and SNP (single nucleotide polymorphism) analyses of DNA sequences. A DNA chip is an array of DNA probes which are hybridized with the compliment sub-sequences in the target sequence. However, conventional DNA chips are showing tendency to be comprised of longer probes and get larger in size to achieve a higher resolution. To shrink the size of DNA chips, the design is considered to be important. To solve this problem, we applied SOM to obtain common features of DNA sequences with small number of probes which efficiently cover the target sequence with sufficient resolution for finding the correct position of SNPs. We evaluated the DNA chips designed by SOM with computer simulations of SNP analyses, changing the length of probes and size of the maps.

Clustering of chromosome fluorescence profiles by self-organising map using chromosome physical models

The clustering by the self-organizing map algorithm of chromosome profiles measured by slit-scan flowcytometer is proposed. Moreover, the physical models of chromosomes have been introduced in order to take into account the rotation of chromosomes in the flowcytometer. The self-organizing map algorithm has been improved so that it can modify the characteristic parameters of chromosome physical models. By this modification, the lengths of chromosomes and the intensity distribution of chromosome fluorescence can be estimated from chromosome profile data measured by the flowcytometer. The estimated lengths of chromosomes have been almost equal to known values of the lengths of chromosomes. The clustering results by the above method have been compared with the clustering results of the same data by K-mean method and an agglomerative hierarchical clustering.

A design method of DNA chips using hierarchical self-organizing maps

We introduce a design method of DNA chips using self organizing maps (SOM) and hierarchical self-organizing maps (H-SOM). DNA chips are powerful tools for sequencings and SNP (single nucleotide polymorphism) analyses of DNA sequences. A DNA chip is an array of DNA probes which can be hybridized with complement subsequences in the target sequence. However, conventional DNA chips are showing tendency to be comprised of large number of long probes and get large in size to achieve high resolution. To shrink the size of DNA chips, design method is considered to be important. To solve this problem, we applied SOM to extract common features of DNA sequences using proper number of probes which efficiently cover the target sequence with sufficient resolutions. Furthermore, H-SOM can design the chip comprised of long probes more efficiently than SOM. We have designed some DNA chips from the sequence data in genome database using our SOM and H-SOM algorithm and evaluated the chips by computer simulations of SNP analyses.

Estimation Method of Solar Cell Temperature Using Meteorological Data in Mega Solar Power Plant

The demands for diagnostics of solar power plants from remote places are increasing. We aim at constructing a diagnostic system which detects faulty modules automatically in mega solar power plants (MSPPs). The system utilizes measured data in power plants which can be obtained comparatively easily, such as string powers or solar irradiances. Solar irradiances are the most important factor which determines the magnitude of power generated in the cell, and cell temperatures are the second most important. However, cell temperatures are not usually measured directly in power plants. In this paper, we propose a new method to estimate the temperature of crystalline silicon solar cells which are most common at present, using measured data such as cell powers, solar irradiances atmospheric temperature, and wind speeds, which can be measured comparatively easily in power plants. We also evaluate our method by comparing estimated cell temperatures and surface temperatures of modules actually measured at Yoshinogari MSPP in Japan.

The adaptive authentication system for behavior biometrics using pareto learning self organizing maps

In this paper, we propose an authentication system which can adapt to the temporal changes of the behavior biometrics with accustoming to the system. We proposed the multi-modal authentication system using Supervised Pareto learning Self Organizing Maps. In this paper, the adaptive authentication system with incremental learning which is applied as the feature of neural networks is developed.

A clustering method of chromosome fluorescence profiles using self organizing map

The clustering method by the self-organizing map algorithm of chromosome profiles measured by slit-scan flowcytometer is proposed. Chromosome profile represents the distribution of the fluorescence intensities along the lengthwise. To examine the performance of the cytometer developed in our laboratory, we made clustering experiments of the measured profiles. We developed a SOM based clustering algorithm using chromosome physical model, which can estimate the chromosome models and rotation angles, but the results were considered to be not complete. For this problem, we propose a new algorithm, which estimate the models and rotation angles step by step. We examine our algorithm using the virtual profiles and profiles measured by our cytometer.

A clustering method of chromosome fluorescence profiles by modified self organizing map

The clustering by the self-organizing map algorithm of chromosome profiles measured by slit-scan flowcytometer is proposed. Moreover, the physical models of chromosomes have been introduced in order to take into account the rotation of chromosomes in the flowcytometer. The self-organizing map algorithm has been improved so that it can modify the characteristic parameters of chromosome physical models. By this modification, the lengths of chromosomes and the intensity distribution of chromosome fluorescence can be estimated from chromosome profile data measured by the flowcytometer. The estimated lengths of chromosomes are almost equal to known values of the lengths of chromosomes. The clustering results by the above method are compared with the clustering results of the same data by the K-mean method and agglomerative hierarchical clustering.