Spherical Self-organizing Map Research Articles

Naturally Fractured Carbonate Reservoir Characterization: A Case Study of a Mature High-Pour Point Oil Field in Hungary

An integrated technical study was conducted for a field development project in west Hungary. This study offers a better solution for estimating petrophysical properties and fracture facies vertically along the well and laterally for 3D static and dynamic models of naturally fractured reservoirs in carbonate rocks. More than 30 wells with 40 years of production history were used in order to build reliable static and dynamic models. The fracture class/facies plays an essential role in the spatial distribution of petrophysical properties during 3D reservoir modeling. It was defined by integrating conventional logs, image logs, drilling parameters, and production or well test data. Three fracture facies are defined as macrofracture (including permeable subseismic fault), microfracture, and host rock. Subsequently, the fracture class’s spatial distribution is guided by seismic attributes of fault likelihood combined with the geological concept of the fault and damage zone. As a result, the established fracture classes along the wells are validated by static and dynamic subsurface data. A spherical self-organizing map (SOM) was also implemented for predicting the open-fracture location in wells having limited subsurface data. Moreover, fracture lateral distribution follows the distribution of the fault zone of the fault core, high damage zone, low damage zone, and host rock. The higher the fault displacement, the wider the damage zone and fault core formed. Macrofractures and microfractures frequently appear around the fault core and high damage zone. While only microfractures are dominantly present in the low damage zones, in contrast, the unfractured class is dominantly distributed in the host-rock area. Also, the lithologies are considered in distributing the fracture class because the rock mechanic properties and the number of fractures are strongly controlled by rock compositions. Once the fracture class is distributed, porosity, permeability, and water saturation are modeled in the three-dimensional (3D) geocellular model. Finally, this fracture class also plays a role as a rock typing for reservoir simulation. The saturation height model is built using the fracture class distribution resulting in the initialization, history-matching process, and production forecast from 20 wells showing excellent quality. As a novelty, this study offers a better understanding of fracture distribution and accelerates the history-matching process with a more confident result of the production forecast. In the absence of advanced technologies, like image logs and production logging (PLT) measurements, this study still effectively assists us in recognizing the fracture presence and its quality in both well-depth interval and 3D spatial space, and successfully guided us in proposing new infill drilling with strong confidence and delivering on the high end of expected results.

Spherical Tree-Structured SOM and Its Application to Hierarchical Clustering

When analyzing high-dimensional data with many elements, a data visualization that maps the data onto a low-dimensional space is often performed. By visualizing the data, humans can intuitively understand the structure of the data in the high-dimensional space. The self-organizing map (SOM) is one such data visualization method. We propose a spherical tree-structured SOM (S-TS-SOM), which speeds up the search for winner nodes and eliminates the unevenness of learning due to the position of the winner nodes by placing the nodes on a sphere and applying the tree search method. In this paper, we confirm that the S-TS-SOM can achieve the same results as a normal spherical SOM while reducing the learning time. In addition, we confirm the granularity of clustering on the tree structure of the S-TS-SOM.

Abstract 2005: CSCs induced from iPSCs indicate the trace of cancer cell culture conditioned media

Abstract Cancer stem cells (CSCs) show drug resistance and could be a source of cancer cells. Although these cells are recently considered more important than ever in cancer studies, the extremely small number of CSCs in actual cancer tissues makes the analysis difficult. Previously, we demonstrated that CSCs could be differentiated from induced pluripotent stem cells (iPSCs) in the presence of cancer cell conditioned media. We have confirmed the conditioned media from various types of cancer cell lines were successful to convert iPSCs into CSCs showing self-renewal property, differentiation potential, and tumorigenicity. To identify their differences between the generated CSCs, gene expression was compared by spherical self-organizing map (sSOM). Comparing with more than 1000 cells including stem cells, induced CSCs, CSCs obtained from cancer tissue or cancer cell line, and cancer cells or cancer tissues, known cancer stem cell markers were found not enough to classify CSCs. In the sSOM analysis, “ideal probe” (IP), which was supposed to characteristically be expressed in each group, was set to nomiate the marker gene. The distance close to the IP depicted some candidate marker genes in each group. The gene expression commonly shared among different types of CSCs was also searched. This IP revealed known cancer related gene Endthelin 1 as a commonly expressed gene. Moreover, when the CSCs were compared with the parental hiPSC and cancer cells, of which conditioned media was used for the conversion, sSOM showed that CSCs were not always differentiated towards their origin, but which genes should be responsible for generating CSCs. These results indicate that the induced CSCs should be the initial state of cancer cells and imply some important role of Endothelin 1 in CSC development. Citation Format: Akimasa Seno, Tomonari Kasai, Heizo Tokutaka, Masaharu Seno. CSCs induced from iPSCs indicate the trace of cancer cell culture conditioned media [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2005.

Commonly expressed genes among cancer stem cells induced from hiPSCs and Obtained from cancer tissues or cell lines

Introduction: Cancer is one of the serious health problems in worldwide. For the development of radical cancer treatment, cancer stem cells (CSCs) are getting an issue of focus. CSCs are thought to be resistant to chemotherapy and cause metastasis. Although it is very important to understand their characters in detail, the knowledge so far is not sufficient due to their low ratio in tumor tissues. Subjects and Methods: We have induced CSCs from induced pluripotent stem cells (iPSCs) culturing in the conditioned media from cancer-derived cells without introducing genes or mutations. These induced CSCs actually have CSCs-like properties of self-renewal, differentiation potential, and tumorigenicity. In this study, their gene expression data are compared with more than 1000 sets of data from normal stem cells, our CSCs, CSCs obtained from cancer tissues or cell lines and cancer cells, which obtained from Gene Expression Omnibus. Results: Although there were no known cancer stem cell markers which can distinguish CSCs from other cell types, clustering with spherical self-organizing map revealed the expression of NTNG1, ABLIM2, DNM3, EDN1, XLOC_001990, and ISY1-RAB43 are significantly high in CSCs. Conclusion: Their expression should help to find CSCs as the markers in the future. Simultaneously, the function of these genes should become important to be clarified.

Analysis of liver damage from radon, X-ray, or alcohol treatments in mice using a self-organizing map.

In our previous studies, we found that low-dose radiation inhibits oxidative stress–induced diseases due to increased antioxidants. Although these effects of low-dose radiation were demonstrated, further research was needed to clarify the effects. However, the analysis of oxidative stress is challenging, especially that of low levels of oxidative stress, because antioxidative substances are intricately involved. Thus, we proposed an approach for analysing oxidative liver damage via use of a self-organizing map (SOM)—a novel and comprehensive technique for evaluating hepatic and antioxidative function. Mice were treated with radon inhalation, irradiated with X-rays, or subjected to intraperitoneal injection of alcohol. We evaluated the oxidative damage levels in the liver from the SOM results for hepatic function and antioxidative substances. The results showed that the effects of low-dose irradiation (radon inhalation at a concentration of up to 2000 Bq/m3, or X-irradiation at a dose of up to 2.0 Gy) were comparable with the effect of alcohol administration at 0.5 g/kg bodyweight. Analysis using the SOM to discriminate small changes was made possible by its ability to ‘learn’ to adapt to unexpected changes. Moreover, when using a spherical SOM, the method comprehensively examined liver damage by radon, X-ray, and alcohol. We found that the types of liver damage caused by radon, X-rays, and alcohol have different characteristics. Therefore, our approaches would be useful as a method for evaluating oxidative liver damage caused by radon, X-rays and alcohol.

Characterization of Gene Expression Patterns among Artificially Developed Cancer Stem Cells Using Spherical Self-Organizing Map.

We performed gene expression microarray analysis coupled with spherical self-organizing map (sSOM) for artificially developed cancer stem cells (CSCs). The CSCs were developed from human induced pluripotent stem cells (hiPSCs) with the conditioned media of cancer cell lines, whereas the CSCs were induced from primary cell culture of human cancer tissues with defined factors (OCT3/4, SOX2, and KLF4). These cells commonly expressed human embryonic stem cell (hESC)/hiPSC-specific genes (POU5F1, SOX2, NANOG, LIN28, and SALL4) at a level equivalent to those of control hiPSC 201B7. The sSOM with unsupervised method demonstrated that the CSCs could be divided into three groups based on their culture conditions and original cancer tissues. Furthermore, with supervised method, sSOM nominated TMED9, RNASE1, NGFR, ST3GAL1, TNS4, BTG2, SLC16A3, CD177, CES1, GDF15, STMN2, FAM20A, NPPB, CD99, MYL7, PRSS23, AHNAK, and LOC152573 genes commonly upregulating among the CSCs compared to hiPSC, suggesting the gene signature of the CSCs.

A Unique Procedure to Identify Cell Surface Markers Through a Spherical Self-Organizing Map Applied to DNA Microarray Analysis.

To identify cell-specific markers, we designed a DNA microarray platform with oligonucleotide probes for human membrane-anchored proteins. Human glioma cell lines were analyzed using microarray and compared with normal and fetal brain tissues. For the microarray analysis, we employed a spherical self-organizing map, which is a clustering method suitable for the conversion of multidimensional data into two-dimensional data and displays the relationship on a spherical surface. Based on the gene expression profile, the cell surface characteristics were successfully mirrored onto the spherical surface, thereby distinguishing normal brain tissue from the disease model based on the strength of gene expression. The clustered glioma-specific genes were further analyzed by polymerase chain reaction procedure and immunocytochemical staining of glioma cells. Our platform and the following procedure were successfully demonstrated to categorize the genes coding for cell surface proteins that are specific to glioma cells. Our assessment demonstrates that a spherical self-organizing map is a valuable tool for distinguishing cell surface markers and can be employed in marker discovery studies for the treatment of cancer.

Spherical Self-Organizing Map Detects MYBL 1 As Candidate Gene for Triple-Negative Breast Cancer

Spherical Self-Organizing Map Detects MYBL 1 As Candidate Gene for Triple-Negative Breast Cancer

A Machine Intelligence Approach to Virtual Ballet Training

This article presents a framework for real-time analysis and visualization of ballet dance movements performed within a Cave Virtual Reality Environment (CAVE). A Kinect sensor captures and extracts dance-based movement features, from which a topology preserved "posture space" is constructed using a spherical self-organizing map (SSOM). Recordings of dance movements are parsed into gestural elements by projection onto the SSOM to form unique trajectories in posture space. Dependencies between postures in a trajectory are modeled using a Markovian empirical transition matrix, which is then used to recognize attempted movements. This allows for quantitative assessment and feedback of a student's performance, delivered using concurrent, localized visualizations together with a performance score based on incremental dynamic time warping (IDTW).

An Approach to Ballet Dance Training through MS Kinect and Visualization in a CAVE Virtual Reality Environment

This article proposes a novel framework for the real-time capture, assessment, and visualization of ballet dance movements as performed by a student in an instructional, virtual reality (VR) setting. The acquisition of human movement data is facilitated by skeletal joint tracking captured using the popular Microsoft (MS) Kinect camera system, while instruction and performance evaluation are provided in the form of 3D visualizations and feedback through a CAVE virtual environment, in which the student is fully immersed. The proposed framework is based on the unsupervised parsing of ballet dance movement into a structuredposture spaceusing the spherical self-organizing map (SSOM). A unique feature descriptor is proposed to more appropriately reflect the subtleties of ballet dance movements, which are represented asgesture trajectoriesthrough posture space on the SSOM. This recognition subsystem is used to identify the category of movement the student is attempting when prompted (by a virtual instructor) to perform a particular dance sequence. The dance sequence is then segmented and cross-referenced against a library of gestural components performed by the teacher. This facilitates alignment and score-based assessment of individual movements within the context of the dance sequence. An immersive interface enables the student to review his or her performance from a number of vantage points, each providing a unique perspective and spatial context suggestive of how the student might make improvements in training. An evaluation of the recognition and virtual feedback systems is presented.

Basic Study on the Classification of Time Series Data Using a Frequency Integrated Spherical Hidden Markov Self Organizing Map

The rapid progress in and the expanding complexity of information and technology systems have made data analysis increasingly relevant. Data having a variety of elements are complex, and making very difficult to evaluate a state of a model from observed data generated probabilistically by the model. To evaluate these hidden states, we propose Spherical-Self Organizing Map (S-SOM) with a Hidden Markov Model (HMM) that infers such hidden states.

Optimized spiral spherical self-organizing map approach to sector analysis – The case of banking

The paper presents a comprehensive treatment of the European Union banking sector in the 2000–2011 period. A wide set of economic indicators is used to derive two-level approach: micro and macro topologies. We use the methodology of the optimized spiral spherical self-organizing map, a version of Kohonen’s Self-Organizing Maps, which overcomes some of the major deficiencies found in standard methods. In dynamic analysis we recover the path of the integration process itself by calculating the dynamics of the countries around the sphere and by examining the measure of dispersion. The simultaneous comparison of the micro and macro levels provides deeper insights into the banking sector integration process. Both models, the micro and macro, unveil the formation of two banking clusters, which, with the exception for Greece, coincides with a division between new and old member states. Both clusters intensify integration within their cluster.

Intuitive volume exploration through spherical self-organizing map and color harmonization

Abstract Finding an appropriate transfer function (TF) for mapping color and opacity values in direct volume rendering (DVR) can be a daunting task. This paper presents a novel approach towards TF generation for DVR, where the traditional low-level color and opacity parameter manipulations are not necessary. The TF generation process is hidden behind a simple and intuitive spherical self-organizing map (SSOM) visualization. The SSOM represents a visual form of the topological relations among the clusters. The user interacts with the SSOM lattice to find interesting regions in the volume. The color and opacity values are generated automatically from the voxel features based on the user׳s perception. We also use harmonic colors to present a visually pleasing result. Due to the independence of SSOM from feature type, our proposed method is flexible in nature and can be integrated with any set of features. The GPU implementation provides real-time volume rendering and fast interaction. Experimental results on several benchmark volume datasets show the effectiveness of our proposed method.

An effective biomedical image retrieval framework in a fuzzy feature space employing Phase Congruency and GeoSOM

This paper presents a detailed study about a biomedical image retrieval framework by extracting Phase Congruency (PC) features from L*a*b* triplets of images (query, target) and representing them in fuzzy feature space. These features correspond to an edge-corner map of the given image. The resulting map is then processed by Scale Invariant Feature Transform (SIFT) to derive keypoints, that are invariant to affine transformations. The ensuing features were vector quantized to build a codebook of keypoints. The codebook was produced using a Spherical Self-Organizing Map (SOM) built with a geodesic data structure termed as GeoSOM. Then keypoints of the query image are mapped with the codebook and their occurrences are counted to formulate a histogram termed as Phase Congruency-based Bag of Keypoints (PC-BoK). This histogram is generated offline for target images and a similarity measure was performed with the query image to yield the nearest match based on a global fuzzy membership function. Exhaustive experiments of the proposed framework named as BIRS (Biomedical Image Retrieval System) were performed on a diverse medical image collection (NBIA, MESSIDOR, DRIVE). Finally, performance of BIRS demonstrates the advantage of the proposed image representation approach in terms of Precision (P)–Recall (R) parameters. Furthermore relative comparison of the proposed scheme with existing feature descriptors depicts improved P–R values. The proposed feature extraction and representation scheme was also robust against quantization errors.

L-shaped PLS analysis of multiple inhibitory activities of adrenergic alpha receptors using ligand and protein matrices

Recently, chemogenomics has been given high attention in pharmaceutical industry. By definition, chemogenomics data is a two-dimensional matrix, where proteins are usually reported as columns and ligands as rows, and where reported values are usually inhibitory activities. In a straightforward manner, it would be the best choice that the chemogenomics matrix is explained by two matrices that each consists of ligand and protein descriptors. Bi-modal PLS follows this concept and several variants have been proposed. Among them, we focus on the L-shaped PLS (LPLS) method and apply it to aminergic G protein-coupled receptor inhibitory activity data in a previous study.In this study, we have devised both of the ligand and protein matrices in the frame of LPLS analysis for four adrenergic alpha receptors. In the ligand matrix, the similarity matrix derived from the Tanimoto similarity value between the pair of inhibitors was employed. As for the protein matrix, the 3D protein pocket was mapped to the spherical self-organizing map sphere. Then, the lipophilic potential value on each node was used as protein descriptors. Thanks to four plots of LPLS, we could easily identify four selective inhibitors and elucidate structural requirements for selectivity.

New description of protein–ligand interactions using a spherical self-organizing map

In a previous report, we studied the mapping ability of the spherical self-organizing map (SSOM). The original 3D structure of the active site of the β2 protein structure was well reproduced by the SSOM. To validate the geometrical transformation and the resulting molecular electrostatic potential (MEP) distribution, the molecular surfaces of 20 β2 ligands were mapped onto the protein SSOM sphere. The MEP values of the two spheres derived from the ligand and the β2 receptor protein were compared. In most cases involving potent ligands, the two spheres had a moderate negative correlation. This indicates that the SSOM approach has excellent potential to represent a complex protein surface as a simple spherical structure.In this study, we perform a quantitative structure–activity relationship (QSAR) study of caspase-3 inhibitors based on the SSOM technique. Initially, the active site of the protein structure ‘caspase-3’ was characterized by the SSOM using the MEP values. Each inhibitor was then projected onto the protein SSOM sphere and the chemical descriptors were derived from the ligand SSOM sphere. The correlation of the chemical descriptors and the inhibitory activities was investigated using the support vector regression (SVR) method. Finally, the important MEP descriptors from the final SVR model were examined. The structural requirements of caspase-3 inhibitors are discussed from the perspectives of both the ligand and protein structures.

A New Method for Mapping the Molecular Surface of a Protein Structure Using a Spherical Self-Organizing Map.

A spherical self-organizing map (SSOM) was applied for mapping the molecular surface of a protein structure on a spherical structure. The active site of the X-ray crystal structure of β2 receptor protein was used for this purpose. After mapping the molecular surface points and assigning the associated molecular electrostatic potential (MEP) values, the original 3D structure of the active site was well reproduced by the SSOM. In order to validate the geometrical transformation and the resulting MEP distribution, the molecular surfaces of twenty β2 ligands were mapped on the established SSOM sphere. The MEP values of two spheres derived from the ligand and the β2 receptor protein were compared. In almost all cases of strong ligands, the two spheres had a moderate negative correlation.

Visualization of Pareto Solutions by Spherical Self-Organizing Map and It’s acceleration on a GPU

In this study, we visualize Pareto-optimum solutions derived from multiple-objective optimization using spherical self-organizing maps (SOMs) that lay out SOM data in three dimensions. There have been a wide range of studies involving plane SOMs where Pareto-optimal solutions are mapped to a plane. However, plane SOMs have an issue that similar data differing in a few specific variables are often placed at far ends of the map, compromising intuitiveness of the visualization. We show in this study that spherical SOMs allow us to find similarities in data otherwise undetectable with plane SOMs. We also implement and evaluate the performance using parallel sphere processing with several GPU environments.

Modeling and predicting binding affinity of phencyclidine‐like compounds using machine learning methods

AbstractMachine learning methods have always been promising in the science and engineering fields, and the use of these methods in chemistry and drug design has advanced especially since the 1990s. In this study, molecular electrostatic potential (MEP) surfaces of phencyclidine‐like (PCP‐like) compounds are modeled and visualized in order to extract features that are useful in predicting binding affinities. In modeling, the Cartesian coordinates of MEP surface points are mapped onto a spherical self‐organizing map (SSOM). The resulting maps are visualized using electrostatic potential (ESP) values. These values also provide features for a prediction system. Support vector machines and partial least‐squares method are used for predicting binding affinities of compounds. Copyright © 2009 John Wiley & Sons, Ltd.

Application of high speed spherical self-organizing map to climate research

Application of high speed spherical self-organizing map to climate research