Decentralized policy learning with partial observation and mechanical constraints for multiperson modeling

Extracting the rules of real-world multi-agent behaviors is a current challenge in various scientific and engineering fields. Biological agents independently have limited observation and mechanical constraints; however, most of the conventional data-driven models ignore such assumptions, resulting in lack of biological plausibility and model interpretability for behavioral analyses. Here we propose sequential generative models with partial observation and mechanical constraints in a decentralized manner, which can model agents' cognition and body dynamics, and predict biologically plausible behaviors. We formulate this as a decentralized multi-agent imitation-learning problem, leveraging binary partial observation and decentralized policy models based on hierarchical variational recurrent neural networks with physical and biomechanical penalties. Using real-world basketball and soccer datasets, we show the effectiveness of our method in terms of the constraint violations, long-term trajectory prediction, and partial observation. Our approach can be used as a multi-agent simulator to generate realistic trajectories using real-world data.

Estimating Counterfactual Treatment Outcomes Over Time in Complex Multiagent Scenarios

Evaluation of intervention in a multiagent system, for example, when humans should intervene in autonomous driving systems and when a player should pass to teammates for a good shot, is challenging in various engineering and scientific fields. Estimating the individual treatment effect (ITE) using counterfactual long-term prediction is practical to evaluate such interventions. However, most of the conventional frameworks did not consider the time-varying complex structure of multiagent relationships and covariate counterfactual prediction. This may lead to erroneous assessments of ITE and difficulty in interpretation. Here, we propose an interpretable, counterfactual recurrent network in multiagent systems to estimate the effect of the intervention. Our model leverages graph variational recurrent neural networks (GVRNNs) and theory-based computation with domain knowledge for the ITE estimation framework based on long-term prediction of multiagent covariates and outcomes, which can confirm the circumstances under which the intervention is effective. On simulated models of an automated vehicle and biological agents with time-varying confounders, we show that our methods achieved lower estimation errors in counterfactual covariates and the most effective treatment timing than the baselines. Furthermore, using real basketball data, our methods performed realistic counterfactual predictions and evaluated the counterfactual passes in shot scenarios.

Manifold alteration between major depressive disorder and healthy control subjects using dynamic mode decomposition in resting-state fMRI data

Background

The World Health Organization has reported that approximately 300 million individuals suffer from the mood disorder known as MDD. Non-invasive measurement techniques have been utilized to reveal the mechanism of MDD, with rsfMRI being the predominant method. The previous functional connectivity and energy landscape studies have shown the difference in the coactivation patterns between MDD and HCs. However, these studies did not consider oscillatory temporal dynamics.

Methods

In this study, the dynamic mode decomposition, a method to compute a set of coherent spatial patterns associated with the oscillation frequency and temporal decay rate, was employed to investigate the alteration of the occurrence of dynamic modes between MDD and HCs. Specifically, The BOLD signals of each subject were transformed into dynamic modes representing coherent spatial patterns and discrete-time eigenvalues to capture temporal variations using dynamic mode decomposition. All the dynamic modes were disentangled into a two-dimensional manifold using t-SNE. Density estimation and density ratio estimation were applied to the two-dimensional manifolds after the two-dimensional manifold was split based on HCs and MDD.

Results

The dynamic modes that uniquely emerged in the MDD were not observed. Instead, we have found some dynamic modes that have shown increased or reduced occurrence in MDD compared with HCs. The reduced dynamic modes were associated with the visual and saliency networks while the increased dynamic modes were associated with the default mode and sensory-motor networks.

Conclusion

To the best of our knowledge, this study showed initial evidence of the alteration of occurrence of the dynamic modes between MDD and HCs. To deepen understanding of how the alteration of the dynamic modes emerges from the structure, it is vital to investigate the relationship between the dynamic modes, cortical thickness, and surface areas.

Dynamic mode decomposition for Koopman spectral analysis of elementary cellular automata

We apply dynamic mode decomposition (DMD) to elementary cellular automata (ECA). Three types of DMD methods are considered, and the reproducibility of the system dynamics and Koopman eigenvalues from observed time series is investigated. While standard DMD fails to reproduce the system dynamics and Koopman eigenvalues associated with a given periodic orbit in some cases, Hankel DMD with delay-embedded time series improves reproducibility. However, Hankel DMD can still fail to reproduce all the Koopman eigenvalues in specific cases. We propose an extended DMD method for ECA that uses nonlinearly transformed time series with discretized Walsh functions and show that it can completely reproduce the dynamics and Koopman eigenvalues. Linear-algebraic backgrounds for the reproducibility of the system dynamics and Koopman eigenvalues are also discussed.

Stable invariant models via Koopman spectra

Weight-tied models have attracted attention in the modern development of neural networks. The deep equilibrium model (DEQ) represents infinitely deep neural networks with weight-tying, and recent studies have shown the potential of this type of approach. DEQs are needed to iteratively solve root-finding problems in training and are built on the assumption that the underlying dynamics determined by the models converge to a fixed point. In this paper, we present the stable invariant model (SIM), a new class of deep models that in principle approximates DEQs under stability and extends the dynamics to more general ones converging to an invariant set (not restricted in a fixed point). The key ingredient in deriving SIMs is a representation of the dynamics with the spectra of the Koopman and Perron-Frobenius operators. This perspective approximately reveals stable dynamics with DEQs and then derives two variants of SIMs. We also propose an implementation of SIMs that can be learned in the same way as feedforward models. We illustrate the empirical performance of SIMs with experiments and demonstrate that SIMs achieve comparative or superior performance against DEQs in several learning tasks.

Predicting behavior through dynamic modes in resting-state fMRI data

Dynamic properties of resting-state functional connectivity (FC) provide rich information on brain-behavior relationships. Dynamic mode decomposition (DMD) has been used as a method to characterize FC dynamics. However, it remains unclear whether dynamic modes (DMs), spatial-temporal coherent patterns computed by DMD, provide information about individual behavioral differences. This study established a methodological approach to predict individual differences in behavior using DMs. Furthermore, we investigated the contribution of DMs within each of seven specific frequency bands (0-0.1,...,0.6-0.7 Hz) for prediction. To validate our approach, we tested whether each of 59 behavioral measures could be predicted by performing multivariate pattern analysis on a Gram matrix, which was created using subject-specific DMs computed from resting-state functional magnetic resonance imaging (rs-fMRI) data of individuals. DMD successfully predicted behavior and outperformed temporal and spatial independent component analysis, which is the conventional data decomposition method for extracting spatial activity patterns. Most of the behavioral measures that were predicted with significant accuracy in a permutation test were related to cognition. We found that DMs within frequency bands <0.2 Hz primarily contributed to prediction and had spatial structures similar to several common resting-state networks. Our results indicate that DMD is efficient in extracting spatiotemporal features from rs-fMRI data.

Koopman spectral analysis of elementary cellular automata

We perform a Koopman spectral analysis of elementary cellular automata (ECA). By lifting the system dynamics using a one-hot representation of the system state, we derive a matrix representation of the Koopman operator as the transpose of the adjacency matrix of the state-transition network. The Koopman eigenvalues are either zero or on the unit circle in the complex plane, and the associated Koopman eigenfunctions can be explicitly constructed. From the Koopman eigenvalues, we can judge the reversibility, determine the number of connected components in the state-transition network, evaluate the period of asymptotic orbits, and derive the conserved quantities for each system. We numerically calculate the Koopman eigenvalues of all rules of ECA on a one-dimensional lattice of 13 cells with periodic boundary conditions. It is shown that the spectral properties of the Koopman operator reflect Wolfram's classification of ECA.

Prediction of Compound Bioactivities Using Heat-Diffusion Equation

Machine learning is expected to improve low throughput and high assay cost in cell-based phenotypic screening. However, it is still a challenge to apply machine learning to achieving sufficiently complex phenotypic screening due to imbalanced datasets, non-linear prediction, and unpredictability of new chemotypes. Here, we developed a prediction model based on the heat-diffusion equation (PM-HDE) to address this issue. The algorithm was verified as feasible for virtual compound screening using biotest data of 946 assay systems registered with PubChem. PM-HDE was then applied to actual screening. Based on supervised learning of the data of about 50,000 compounds from biological phenotypic screening with motor neurons derived from ALS-patient-induced pluripotent stem cells, virtual screening of >1.6 million compounds was implemented. We confirmed that PM-HDE enriched the hit compounds and identified new chemotypes. This prediction model could overcome the inflexibility in machine learning, and our approach could provide a novel platform for drug discovery.

•

Prediction model based on heat-diffusion equation (PM-HDE) was constructed

•

PM-HDE succeeded in increasing the hit ratio and identifying potent compounds

•

PM-HDE discovered new chemotypes in compound evaluation with an ALS-patient iPSC panel

•

PM-HDE could represent an algorithm for future drug discovery with AI

There remain many intractable diseases with no treatment available, including amyotrophic lateral sclerosis (ALS), for which the development of a cure is crucial. However, compound screening for drug development demands time, energy, and cost, and therefore artificial intelligence (AI) is expected to improve the efficiency of drug discovery. We built a novel machine-learning algorithm to predict hit compounds in compound screening using the heat-diffusion equation (HDE). This prediction model harbors the potential to solve issues that have been challenging for conventional machine learning and to exhibit accurate performance leading to the discovery of new drugs. In fact, the HDE model predicted hits with new chemotypes among millions of compounds for ALS therapeutics using a panel of large numbers of ALS patient-derived induced pluripotent stem cell models (ALS-patient iPSC panel). This algorithm could contribute to the acceleration and development of future drug discoveries using AI.

Cognition and interpersonal coordination of patients with schizophrenia who have sports habits

Team sports activities are effective for improving the negative symptoms and cognitive functions in patients with schizophrenia. However, the interpersonal coordination during the sports and visual cognition of patients with schizophrenia who have team sports habits are unknown. The main objectives of this study were to test two hypotheses: first, patients with schizophrenia perform the skill requiring ball passing and receiving worse than healthy controls; and second, the patients will be impaired in these functionings in accordance with the previous studies regarding schizophrenia in general. Twelve patients with schizophrenia and 15 healthy controls, who had habits in football, participated in this study. The participants performed three conventional cognitive tests and a 3-vs-1 ball possession task to evaluate their interpersonal coordination. The results showed that in the 3-vs-1 possession task, the displacement in the pass angle for the patients was significantly smaller than that for the control. The recall in the complex figure test, the performance in the trail making test, and that in the five-choice reaction task for the patients were worse than those for the control. Moreover, we found the significant partial correlations in the patients between the extradimensional shift error and the pass angle as well as between the time in the trail making test and the displacement in the pass angle, whereas there was no significant correlation in the control group. This study clarified the impaired interpersonal coordination during team sports and the visual cognition of patients with schizophrenia who have team sports habits.

Whole-Genome Sequencing of the NARO World Rice Core Collection (WRC) as the Basis for Diversity and Association Studies

Genebanks provide access to diverse materials for crop improvement. To utilize and evaluate them effectively, core collections, such as the World Rice Core Collection (WRC) in the Genebank at the National Agriculture and Food Research Organization, have been developed. Because the WRC consists of 69 accessions with a high degree of genetic diversity, it has been used for >300 projects. To allow deeper investigation of existing WRC data and to further promote research using Genebank rice accessions, we performed whole-genome resequencing of these 69 accessions, examining their sequence variation by mapping against the Oryza sativa ssp. japonica Nipponbare genome. We obtained a total of 2,805,329 single nucleotide polymorphisms (SNPs) and 357,639 insertion-deletions. Based on the principal component analysis and population structure analysis of these data, the WRC can be classified into three major groups. We applied TASUKE, a multiple genome browser to visualize the different WRC genome sequences, and classified haplotype groups of genes affecting seed characteristics and heading date. TASUKE thus provides access to WRC genotypes as a tool for reverse genetics. We examined the suitability of the compact WRC population for genome-wide association studies (GWASs). Heading date, affected by a large number of quantitative trait loci (QTLs), was not associated with known genes, but several seed-related phenotypes were associated with known genes. Thus, for QTLs of strong effect, the compact WRC performed well in GWAS. This information enables us to understand genetic diversity in 37,000 rice accessions maintained in the Genebank and to find genes associated with different phenotypes. The sequence data have been deposited in DNA Data Bank of Japan Sequence Read Archive (DRA) (Supplementary Table S1).

Effects of Sulfonylureas on Periodontopathic Bacteria-Induced Inflammation

Interleukin-1β (IL-1β) is an inflammatory cytokine produced by monocytes/macrophages and is closely associated with periodontal diseases. The NLRP3 inflammasome is involved in IL-1β activation through pro-IL-1β processing and pyroptotic cell death in bacterial infection. Recently, glyburide, a hypoglycemic sulfonylurea, has been reported to reduce IL-1β activation by suppressing activation of the NLRP3 inflammasome. Therefore, we evaluated the possibility of targeting the NLRP3 inflammasome pathway by glyburide to suppress periodontal pathogen-induced inflammation. THP-1 cells (a human monocyte cell line) were differentiated to macrophage-like cells by treatment with phorbol 12-myristate 13-acetate and stimulated by periodontopathic bacteria, Porphyromonas gingivalis, Aggregatibacter actinomycetemcomitans, or Fusobacterium nucleatum, in the presence of glyburide. IL-1β and caspase-1 expression in the cells and culture supernatants were analyzed by Western blotting and enzyme-linked immunosorbent assay, and cell death was analyzed by lactate dehydrogenase assay. Stimulation of THP-1 macrophage-like cells with every periodontopathic bacteria induced IL-1β secretion without cell death, which was suppressed by the NLRP3 inhibitor, MCC950, and caspase-1 inhibitor, z-YVAD-FMK. Glyburide treatment suppressed IL-1β expression in culture supernatants and enhanced intracellular IL-1β expression, suggesting that glyburide may have inhibited IL-1β secretion. Subsequently, a periodontitis rat model was generated by injecting periodontal bacteria into the gingiva, which was analyzed histologically. Oral administration of glyburide significantly suppressed the infiltration of inflammatory cells and the number of osteoclasts in the alveolar bone compared with the control. In addition to glyburide, glimepiride was shown to suppress the release of IL-1β from THP-1 macrophage-like cells, whereas other sulfonylureas (tolbutamide and gliclazide) or other hypoglycemic drugs belonging to the biguanide family, such as metformin, failed to suppress IL-1β release. Our results suggest that pharmacological targeting of the NLRP3 pathway may be a strategy for suppressing periodontal diseases.

Physically-interpretable classification of biological network dynamics for complex collective motions

Understanding biological network dynamics is a fundamental issue in various scientific and engineering fields. Network theory is capable of revealing the relationship between elements and their propagation; however, for complex collective motions, the network properties often transiently and complexly change. A fundamental question addressed here pertains to the classification of collective motion network based on physically-interpretable dynamical properties. Here we apply a data-driven spectral analysis called graph dynamic mode decomposition, which obtains the dynamical properties for collective motion classification. Using a ballgame as an example, we classified the strategic collective motions in different global behaviours and discovered that, in addition to the physical properties, the contextual node information was critical for classification. Furthermore, we discovered the label-specific stronger spectra in the relationship among the nearest agents, providing physical and semantic interpretations. Our approach contributes to the understanding of principles of biological complex network dynamics from the perspective of nonlinear dynamical systems.

P1520 Aortic flow reversal caused by aortic regurgitation deteriorates renal function

Background

Chronic kidney disease is a growing public health problem. Renal dysfunction is known as a strong risk factor for cardiovascular disease and end-stage renal failure. The presence of pan-diastolic flow reversal in the abdominal aorta is a very specific sign of severe aortic regurgitation (AR). A higher aortic reverse/forward flow ratio is associated with lower intrarenal forward flow. However, the influence of AR on renal function has been poorly understood. We hypothesized that the aortic flow reversal reduces the renal artery forward flow and accordingly leads to renal dysfunction in patients with severe AR.

Methods

The study consisted of 21 consecutive patients (mean age 69 ± 11 years) with severe AR who underwent aortic valve replacement (AVR). We compared echocardiographic indices and the glomerular filtration rate (GFR) before and 603 ± 541 days after AVR.

Results

Blood pressure was 122 ± 16/54 ± 8 mmHg before AVR and 123 ± 16/76 ± 11 mmHg after AVR. After AVR, left ventricular (LV) end-diastolic dimension decreased from 57 ± 9 to 44 ± 5 mm and LV ejection fraction increased from 58 ± 12 to 60 ± 11 %. Estimated GFR significantly increased from 62.9 ± 18.9 to 71.8 ± 18.1 mL/min per 1.73 m2 after AVR (p = 0.003).

Conclusions An increase in aortic flow reversal caused by severe AR reduces forward flow into the kidney and thereby deteriorates renal function. This study demonstrated a key mediating role of central hemodynamic factors, particularly an exaggerated aortic flow reversal in renal dysfunction and severe AR.

Dynamic mode decomposition in vector-valued reproducing kernel Hilbert spaces for extracting dynamical structure among observables

Understanding nonlinear dynamical systems (NLDSs) is challenging in a variety of engineering and scientific fields. Dynamic mode decomposition (DMD), which is a numerical algorithm for the spectral analysis of Koopman operators, has been attracting attention as a way of obtaining global modal descriptions of NLDSs without requiring explicit prior knowledge. However, since existing DMD algorithms are in principle formulated based on the concatenation of scalar observables, it is not directly applicable to data with dependent structures among observables, which take, for example, the form of a sequence of graphs. In this paper, we formulate Koopman spectral analysis for NLDSs with structures among observables and propose an estimation algorithm for this problem. This method can extract and visualize the underlying low-dimensional global dynamics of NLDSs with structures among observables from data, which can be useful in understanding the underlying dynamics of such NLDSs. To this end, we first formulate the problem of estimating spectra of the Koopman operator defined in vector-valued reproducing kernel Hilbert spaces, and then develop an estimation procedure for this problem by reformulating tensor-based DMD. As a special case of our method, we propose the method named as Graph DMD, which is a numerical algorithm for Koopman spectral analysis of graph dynamical systems, using a sequence of adjacency matrices. We investigate the empirical performance of our method by using synthetic and real-world data.

Automatically recognizing strategic cooperative behaviors in various situations of a team sport

Understanding multi-agent cooperative behavior is challenging in various scientific and engineering domains. In some cases, such as team sports, many cooperative behaviors can be visually categorized and labeled manually by experts. However, these actions which are manually categorized with the same label based on its function have low spatiotemporal similarity. In other words, it is difficult to find similar and different structures of the motions with the same and different labels, respectively. Here, we propose an automatic recognition system for strategic cooperative plays, which are the minimal, basic, and diverse plays in a ball game. Using player’s moving distance, geometric information, and distances among players, the proposed method accurately discriminated not only the cooperative plays in a primary area, i.e., near the ball, but also those distant from a primary area. We also propose a method to classify more detailed types of cooperative plays in various situations. The proposed framework, which sheds light on inconspicuous players to play important roles, could have a potential to detect well-defined and labeled cooperative behaviors.

Translocator protein 18kDa antagonist ameliorates stress-induced stool abnormality and abdominal pain in rodent stress models

BACKGROUND

Irritable bowel syndrome (IBS) is a functional gastrointestinal (GI) disorder characterized by abdominal pain and abnormal bowel habits, both of which are exacerbated by psychological stress. The translocator protein 18kDa (TSPO) is a marker of reactive gliosis in a number of central nervous system (CNS) diseases and responsible for many cellular functions, including neurosteroidogenesis. Although it has been reported that psychological stress disturbs neurosteroids levels, the pathophysiological relevance of TSPO in IBS is poorly understood.

METHODS

We examined the effects of a TSPO antagonist, ONO-2952, on stress-induced stool abnormality and abdominal pain in rats, and on anxiety-related behavior induced by cholecystokinin.

KEY RESULTS

Oral administration of ONO-2952 attenuated stress-induced defecation and rectal hyperalgesia in rats with an efficacy equivalent to that of a 5-HT₃ receptor antagonist. In addition, ONO-2952 suppressed cholecystokinin-induced anxiety-like behavior with an efficacy equivalent to that of psychotropic drugs. On the other hand, ONO-2952 did not affect spontaneous defecation, gastrointestinal transit, visceral nociceptive threshold, and neurosteroid production in non-stressed rats even at a dose 10 times higher than its effective dose in the stress models.

CONCLUSIONS AND INFERENCES

These results suggest that TSPO antagonism results in antistress action, and that ONO-2952 is a promising candidate for IBS without side effects associated with current treatment.

Highly biocompatible super-resolution fluorescence imaging using the fast photoswitching fluorescent protein Kohinoor and SPoD-ExPAN with Lp-regularized image reconstruction

Far-field super-resolution fluorescence microscopy has enabled us to visualize live cells in great detail and with an unprecedented resolution. However, the techniques developed thus far have required high-power illumination (102-106 W/cm2), which leads to considerable phototoxicity to live cells and hampers time-lapse observation of the cells. In this study we show a highly biocompatible super-resolution microscopy technique that requires a very low-power illumination. The present technique combines a fast photoswitchable fluorescent protein, Kohinoor, with SPoD-ExPAN (super-resolution by polarization demodulation/excitation polarization angle narrowing). With this technique, we successfully observed Kohinoor-fusion proteins involving vimentin, paxillin, histone and clathrin expressed in HeLa cells at a spatial resolution of 70-80 nm with illumination power densities as low as ~1 W/cm2 for both excitation and photoswitching. Furthermore, although the previous SPoD-ExPAN technique used L1-regularized maximum-likelihood calculations to reconstruct super-resolved images, we devised an extension to the Lp-regularization to obtain super-resolved images that more accurately describe objects at the specimen plane. Thus, the present technique would significantly extend the applicability of super-resolution fluorescence microscopy for live-cell imaging.

Subspace dynamic mode decomposition for stochastic Koopman analysis

The analysis of nonlinear dynamical systems based on the Koopman operator is attracting attention in various applications. Dynamic mode decomposition (DMD) is a data-driven algorithm for Koopman spectral analysis, and several variants with a wide range of applications have been proposed. However, popular implementations of DMD suffer from observation noise on random dynamical systems and generate inaccurate estimation of the spectra of the stochastic Koopman operator. In this paper, we propose subspace DMD as an algorithm for the Koopman analysis of random dynamical systems with observation noise. Subspace DMD first computes the orthogonal projection of future snapshots to the space of past snapshots and then estimates the spectra of a linear model, and its output converges to the spectra of the stochastic Koopman operator under standard assumptions. We investigate the empirical performance of subspace DMD with several dynamical systems and show its utility for the Koopman analysis of random dynamical systems.

Efficient Generalized Fused Lasso and Its Applications

Generalized fused lasso (GFL) penalizes variables with l 1 norms based both on the variables and their pairwise differences. GFL is useful when applied to data where prior information is expressed using a graph over the variables. However, the existing GFL algorithms incur high computational costs and do not scale to high-dimensional problems. In this study, we propose a fast and scalable algorithm for GFL. Based on the fact that fusion penalty is the Lovász extension of a cut function, we show that the key building block of the optimization is equivalent to recursively solving graph-cut problems. Thus, we use a parametric flow algorithm to solve GFL in an efficient manner. Runtime comparisons demonstrate a significant speedup compared to existing GFL algorithms. Moreover, the proposed optimization framework is very general; by designing different cut functions, we also discuss the extension of GFL to directed graphs. Exploiting the scalability of the proposed algorithm, we demonstrate the applications of our algorithm to the diagnosis of Alzheimer’s disease (AD) and video background subtraction (BS). In the AD problem, we formulated the diagnosis of AD as a GFL regularized classification. Our experimental evaluations demonstrated that the diagnosis performance was promising. We observed that the selected critical voxels were well structured, i.e., connected, consistent according to cross validation, and in agreement with prior pathological knowledge. In the BS problem, GFL naturally models arbitrary foregrounds without predefined grouping of the pixels. Even by applying simple background models, e.g., a sparse linear combination of former frames, we achieved state-of-the-art performance on several public datasets.

Proton pump inhibitor after endoscopic resection for esophageal squamous cell cancer: multicenter prospective randomized controlled trial

BACKGROUND

Whether proton pump inhibitors (PPIs) relieve heartburn or precordial pain after endoscopic resection (ER) for esophageal squamous cell carcinoma (ESCC) remains unclear. The aim of this study was to investigate the efficacy of PPI therapy for these symptoms after ER for ESCC.

METHODS

We conducted a multicenter prospective randomized controlled trial among 15 hospitals in Japan. In total, 229 patients with cT1a ESCC were randomly assigned to receive PPI therapy for 5 weeks after ER (the PPI group, n = 115) or follow-up without PPI therapy (the non-PPI group, n = 114). The primary end point was the incidence of gastroesophageal reflux disease (GERD)-like symptoms after ER from a self-reported questionnaire (Frequency Scale for Symptoms of GERD). Secondary end points were ulcer healing rate at 5 weeks, incidence of pain, improvement rate of symptoms in those who started PPI therapy because of GERD-like symptoms in the non-PPI group, and adverse events.

RESULTS

No significant difference was observed in the incidence of GERD-like symptoms after ER between the non-PPI and PPI groups (30 % vs 34 %, respectively). No significant differences were observed in the ulcer healing rate at 5 weeks (84 % vs 85 %) and incidence of pain within 1 week (36 % vs 45 %). In nine of ten patients (90 %) who started PPI therapy because of GERD-like symptoms in the non-PPI group, PPI administration relieved GERD-like symptoms. No adverse events related to PPI administration were observed.

CONCLUSION

PPI therapy is not efficacious in reducing symptoms and did not promote healing of ulcers in patients undergoing ER for ESCC.

A Novel Continuous and Structural VAR Modeling Approach and Its Application to Reactor Noise Analysis

A vector autoregressive model in discrete time domain (DVAR) is often used to analyze continuous time, multivariate, linear Markov systems through their observed time series data sampled at discrete timesteps. Based on previous studies, the DVAR model is supposed to be a noncanonical representation of the system, that is, it does not correspond to a unique system bijectively. However, in this article, we characterize the relations of the DVAR model with its corresponding Structural Vector AR (SVAR) and Continuous Time Vector AR (CTVAR) models through a finite difference method across continuous and discrete time domain. We further clarify that the DVAR model of a continuous time, multivariate, linear Markov system is canonical under a highly generic condition. Our analysis shows that we can uniquely reproduce its SVAR and CTVAR models from the DVAR model. Based on these results, we propose a novel Continuous and Structural Vector Autoregressive (CSVAR) modeling approach to derive the SVAR and the CTVAR models from their DVAR model empirically derived from the observed time series of continuous time linear Markov systems. We demonstrate its superior performance through some numerical experiments on both artificial and real-world data.

Toxicity prediction from toxicogenomic data based on class association rule mining

While the recent advent of new technologies in biology such as DNA microarray and next-generation sequencer has given researchers a large volume of data representing genome-wide biological responses, it is not necessarily easy to derive knowledge that is accurate and understandable at the same time. In this study, we applied the Classification Based on Association (CBA) algorithm, one of the class association rule mining techniques, to the TG-GATEs database, where both toxicogenomic and toxicological data of more than 150 compounds in rat and human are stored. We compared the generated classifiers between CBA and linear discriminant analysis (LDA) and showed that CBA is superior to LDA in terms of both predictive performances (accuracy: 83% for CBA vs. 75% for LDA, sensitivity: 82% for CBA vs. 72% for LDA, specificity: 85% for CBA vs. 75% for LDA) and interpretability.

Efficient network-guided multi-locus association mapping with graph cuts

MOTIVATION

As an increasing number of genome-wide association studies reveal the limitations of the attempt to explain phenotypic heritability by single genetic loci, there is a recent focus on associating complex phenotypes with sets of genetic loci. Although several methods for multi-locus mapping have been proposed, it is often unclear how to relate the detected loci to the growing knowledge about gene pathways and networks. The few methods that take biological pathways or networks into account are either restricted to investigating a limited number of predetermined sets of loci or do not scale to genome-wide settings.

RESULTS

We present SConES, a new efficient method to discover sets of genetic loci that are maximally associated with a phenotype while being connected in an underlying network. Our approach is based on a minimum cut reformulation of the problem of selecting features under sparsity and connectivity constraints, which can be solved exactly and rapidly. SConES outperforms state-of-the-art competitors in terms of runtime, scales to hundreds of thousands of genetic loci and exhibits higher power in detecting causal SNPs in simulation studies than other methods. On flowering time phenotypes and genotypes from Arabidopsis thaliana, SConES detects loci that enable accurate phenotype prediction and that are supported by the literature.

AVAILABILITY

Code is available at http://webdav.tuebingen.mpg.de/u/karsten/Forschung/scones/.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Active learning for noisy oracle via density power divergence

The accuracy of active learning is critically influenced by the existence of noisy labels given by a noisy oracle. In this paper, we propose a novel pool-based active learning framework through robust measures based on density power divergence. By minimizing density power divergence, such as β-divergence and γ-divergence, one can estimate the model accurately even under the existence of noisy labels within data. Accordingly, we develop query selecting measures for pool-based active learning using these divergences. In addition, we propose an evaluation scheme for these measures based on asymptotic statistical analyses, which enables us to perform active learning by evaluating an estimation error directly. Experiments with benchmark datasets and real-world image datasets show that our active learning scheme performs better than several baseline methods.

Separation of stationary and non-stationary sources with a generalized eigenvalue problem

Non-stationary effects are ubiquitous in real world data. In many settings, the observed signals are a mixture of underlying stationary and non-stationary sources that cannot be measured directly. For example, in EEG analysis, electrodes on the scalp record the activity from several sources located inside the brain, which one could only measure invasively. Discerning stationary and non-stationary contributions is an important step towards uncovering the mechanisms of the data generating system. To that end, in Stationary Subspace Analysis (SSA), the observed signal is modeled as a linear superposition of stationary and non-stationary sources, where the aim is to separate the two groups in the mixture. In this paper, we propose the first SSA algorithm that has a closed form solution. The novel method, Analytic SSA (ASSA), is more than 100 times faster than the state-of-the-art, numerically stable, and guaranteed to be optimal when the covariance between stationary and non-stationary sources is time-constant. In numerical simulations on wide range of settings, we show that our method yields superior results, even for signals with time-varying group-wise covariance. In an application to geophysical data analysis, ASSA extracts meaningful components that shed new light on the Pi 2 pulsations of the geomagnetic field.

Role of macrophage colony-stimulating factor (M-CSF)-dependent macrophages in gastric ulcer healing in mice

We examined the role of macrophage colony-stimulating factor (M-CSF)-dependent macrophages in the healing of gastric ulcers in mice. Male M-CSF-deficient (op/op) and M-CSF-expressing heterozygote (+/?) mice were used. Gastric ulcers were induced by thermal cauterization under ether anesthesia, and healing was observed for 14 days after ulceration. The numbers of macrophages and microvessels in the gastric mucosa were determined immunohistochemically with anti-CD68 and anti-CD31 antibodies, respectively. Expression of tumor necrosis factor (TNF)-α, cyclooxygenase (COX)-2, and vascular endothelial growth factor (VEGF) mRNA was determined via real-time reverse transcription-polymerase chain reaction (RT-PCR), and the mucosal content of prostaglandin (PG) E(2) was determined via enzyme immunoassay on day 10 after ulceration. The healing of gastric ulcers was significantly delayed in op/op mice compared with +/? mice. Further, significantly fewer macrophages were observed in the normal gastric mucosa of op/op mice than in +/? mice. Ulcer induction caused a marked accumulation of macrophages around the ulcer base in +/? mice, but this response was attenuated in op/op mice. The mucosal PGE(2) content as well as the expression of COX-2, VEGF, and TNF-α mRNA were all upregulated in the ulcerated area of +/? mice but significantly suppressed in op/op mice. The degree of vascularization in the ulcerated area was significantly lower in op/op mice than in +/? mice. Taken together, these results suggest that M-CSF-dependent macrophages play an important role in the healing of gastric ulcers, and that this action may be associated with angiogenesis promoted by upregulation of COX-2/PGE(2) production.

[Surgical treatment of postinfarction ventricular septal rupture]

BACKGROUND

Postinfarction ventricular septal rupture (VSR) is a lethal complication with high mortality. The aim of this study was to evaluate our surgical strategy and results of VSR.

PATIENTS AND METHODS

Between 1996 and 2008, 13 consecutive patients underwent operation for VSR at our hospital. All patients required emergent operation because of severe cardiogenic shock. Surgical procedure consisted of endocardial patch repair with infarct exclusion, so called "Komeda-David operation". In patients with multiple coronary artery disease, myocardial revascularization was done simultaneously.

RESULTS

These patients were divided into 2 groups according to the location of VSR. There were 9 patients of anterior VSR. Two of them could not be weaned from cardiopulmonary bypass and died of severe low output syndrome (LOS) at early postoperative period. The site of infarction in both patients was broad anteroseptal region including right ventricle. On the other hand, there were 4 patients of inferior VSP. Two of these patients were lost due to LOS. One patient was complicated with left ventricular free wall rupture. In another patient, infarction was extended proximally toward the mitral annulus and papillary muscles. Both cardiopulmonary bypass time and aortic crossclamp time were significantly longer in inferior VSR than in anterior region. There was no late death in 2 groups.

CONCLUSIONS

Despite improvements of surgical procedures, such as infarct exclusion technique, the operative mortality remains high in cases with broad infarction and/or right ventricular infarction. In these particular circumstances, in should be mandatory to consider the optimal timing of operation and the modification of surgical technique itself.

Prospective randomized controlled study on the effects of perioperative administration of a neutrophil elastase inhibitor to patients undergoing video-assisted thoracoscopic surgery for thoracic esophageal cancer

Sivelestat sodium hydrate (Ono Pharmaceutical Co., Osaka, Japan) is a selective inhibitor of neutrophil elastase (NE) and is effective in reducing acute lung injury associated with systemic inflammatory response syndrome (SIRS). We conducted a prospective randomized controlled study to investigate the efficacy of perioperative administration of sivelestat sodium hydrate to prevent postoperative acute lung injury in patients undergoing thoracoscopic esophagectomy and radical lymphadenectomy. Twenty-two patients with thoracic esophageal cancer underwent video-assisted thoracoscopic esophagectomy with extended lymph node dissection in our institution between April 2007 and November 2008. Using a double-blinded method, these patients were randomly assigned to one of two groups preoperatively. The active treatment group received sivelestat sodium hydrate intravenously for 72 hours starting at the beginning of surgery (sivelestat-treated group; n= 11), while the other group received saline (control group; n= 11). All patients were given methylprednisolone immediately before surgery. Postoperative clinical course was compared between the two groups. Two patients (one in each group) were discontinued from the study during the postoperative period because of surgery-related complications. Of the remaining 20 patients, 2 patients who developed pneumonia within a week after surgery were excluded from some laboratory analyses, so data from 18 patients (9 patients in each group) were analyzed based on the arterial oxygen pressure/fraction of inspired oxygen ratio, white blood cell count, serum C-reactive protein level, plasma cytokine levels, plasma NE level, and markers of alveolar type II epithelial cells. In the current study, the incidence of postoperative morbidity did not differ between the two groups. The median duration of SIRS in the sivelestat-treated group was significantly shorter than that in the control group: 17 (range 9-36) hours versus 49 (15-60) hours, respectively (P= 0.009). Concerning the parameters used for the diagnosis of SIRS, the median heart rates on postoperative day (POD) 2 were significantly lower in the sivelestat-treated group than in the control group (P= 0.007). The median arterial oxygen pressure/fraction of inspired oxygen ratio of the sivelestat-treated group were significantly higher than those of the control group on POD 1 and POD 7 (POD 1: 372.0 [range 284.0-475.0] vs 322.5 [243.5-380.0], respectively, P= 0.040; POD 7: 377.2 [339.5-430.0] vs 357.6 [240.0-392.8], P= 0.031). Postoperative white blood cell counts, serum C-reactive protein levels, plasma interleukin-1beta, tumor necrosis factor-alpha levels, and plasma NE levels did not differ significantly between the two groups at any point during the postoperative course, nor did serum Krebs von den Lungen 6, surfactant protein-A, or surfactant protein-D levels, which were used as markers of alveolar type II epithelial cells to evaluate the severity of lung injury. Plasma interleukin-8 levels were significantly lower in the sivelestat-treated group than in the control group on POD 3 (P= 0.040). In conclusion, perioperative administration of sivelestat sodium hydrate (starting at the beginning of surgery) mitigated postoperative hypoxia, partially suppressed postoperative hypercytokinemia, shortened the duration of SIRS, and stabilized postoperative circulatory status after thoracoscopic esophagectomy.

Development of a higher power intermediate-frequency magnetic field exposure system for in vitro studies

In a previous article we developed an in vitro 23 kHz magnetic field (MF) exposure system that generated an MF of 532 microT(rms). Using this system, the biological effects of 23 kHz MFs on cell functions have been reported. To further clarify the biological effect of intermediate-frequency (IF) MFs and investigate the dose-response relationship in cell lines, an exposure system that generates stronger MFs is required. To meet this requirement, we developed a 6.25 mT(rms) MF exposure system for in vitro study. This level is 1000 times the reference level for the general public in the ICNIRP guidelines. This system provides an MF of 6.25 mT(rms) at 23 kHz with a uniformity within +/-5%. To verify that in vitro experimental conditions are maintained, we examined the temperature, environmental MF, and MF leakage for a sham exposure system. In addition, we examined the harmonics, coil shape, and heat generated in the medium by the high-strength MF. As a result, it was confirmed that this system can be used to evaluate the biological effects of IF MFs. This article presents the design and successful construction of the in vitro exposure system.