An ensemble model of QSAR tools for regulatory risk assessment.

PubMed

Pradeep, Prachi; Povinelli, Richard J; White, Shannon; Merrill, Stephen J

2016-01-01

Quantitative structure activity relationships (QSARs) are theoretical models that relate a quantitative measure of chemical structure to a physical property or a biological effect. QSAR predictions can be used for chemical risk assessment for protection of human and environmental health, which makes them interesting to regulators, especially in the absence of experimental data. For compatibility with regulatory use, QSAR models should be transparent, reproducible and optimized to minimize the number of false negatives. In silico QSAR tools are gaining wide acceptance as a faster alternative to otherwise time-consuming clinical and animal testing methods. However, different QSAR tools often make conflicting predictions for a given chemical and may also vary in their predictive performance across different chemical datasets. In a regulatory context, conflicting predictions raise interpretation, validation and adequacy concerns. To address these concerns, ensemble learning techniques in the machine learning paradigm can be used to integrate predictions from multiple tools. By leveraging various underlying QSAR algorithms and training datasets, the resulting consensus prediction should yield better overall predictive ability. We present a novel ensemble QSAR model using Bayesian classification. The model allows for varying a cut-off parameter that allows for a selection in the desirable trade-off between model sensitivity and specificity. The predictive performance of the ensemble model is compared with four in silico tools (Toxtree, Lazar, OECD Toolbox, and Danish QSAR) to predict carcinogenicity for a dataset of air toxins (332 chemicals) and a subset of the gold carcinogenic potency database (480 chemicals). Leave-one-out cross validation results show that the ensemble model achieves the best trade-off between sensitivity and specificity (accuracy: 83.8 % and 80.4 %, and balanced accuracy: 80.6 % and 80.8 %) and highest inter-rater agreement [kappa ( κ ): 0.63 and 0.62] for both the datasets. The ROC curves demonstrate the utility of the cut-off feature in the predictive ability of the ensemble model. This feature provides an additional control to the regulators in grading a chemical based on the severity of the toxic endpoint under study.

An ensemble model of QSAR tools for regulatory risk assessment

DOE PAGES

Pradeep, Prachi; Povinelli, Richard J.; White, Shannon; ...

2016-09-22

Quantitative structure activity relationships (QSARs) are theoretical models that relate a quantitative measure of chemical structure to a physical property or a biological effect. QSAR predictions can be used for chemical risk assessment for protection of human and environmental health, which makes them interesting to regulators, especially in the absence of experimental data. For compatibility with regulatory use, QSAR models should be transparent, reproducible and optimized to minimize the number of false negatives. In silico QSAR tools are gaining wide acceptance as a faster alternative to otherwise time-consuming clinical and animal testing methods. However, different QSAR tools often make conflictingmore » predictions for a given chemical and may also vary in their predictive performance across different chemical datasets. In a regulatory context, conflicting predictions raise interpretation, validation and adequacy concerns. To address these concerns, ensemble learning techniques in the machine learning paradigm can be used to integrate predictions from multiple tools. By leveraging various underlying QSAR algorithms and training datasets, the resulting consensus prediction should yield better overall predictive ability. We present a novel ensemble QSAR model using Bayesian classification. The model allows for varying a cut-off parameter that allows for a selection in the desirable trade-off between model sensitivity and specificity. The predictive performance of the ensemble model is compared with four in silico tools (Toxtree, Lazar, OECD Toolbox, and Danish QSAR) to predict carcinogenicity for a dataset of air toxins (332 chemicals) and a subset of the gold carcinogenic potency database (480 chemicals). Leave-one-out cross validation results show that the ensemble model achieves the best trade-off between sensitivity and specificity (accuracy: 83.8 % and 80.4 %, and balanced accuracy: 80.6 % and 80.8 %) and highest inter-rater agreement [kappa (κ): 0.63 and 0.62] for both the datasets. The ROC curves demonstrate the utility of the cut-off feature in the predictive ability of the ensemble model. In conclusion, this feature provides an additional control to the regulators in grading a chemical based on the severity of the toxic endpoint under study.« less

Potentialities of ensemble strategies for flood forecasting over the Milano urban area

NASA Astrophysics Data System (ADS)

Ravazzani, Giovanni; Amengual, Arnau; Ceppi, Alessandro; Homar, Víctor; Romero, Romu; Lombardi, Gabriele; Mancini, Marco

2016-08-01

Analysis of ensemble forecasting strategies, which can provide a tangible backing for flood early warning procedures and mitigation measures over the Mediterranean region, is one of the fundamental motivations of the international HyMeX programme. Here, we examine two severe hydrometeorological episodes that affected the Milano urban area and for which the complex flood protection system of the city did not completely succeed. Indeed, flood damage have exponentially increased during the last 60 years, due to industrial and urban developments. Thus, the improvement of the Milano flood control system needs a synergism between structural and non-structural approaches. First, we examine how land-use changes due to urban development have altered the hydrological response to intense rainfalls. Second, we test a flood forecasting system which comprises the Flash-flood Event-based Spatially distributed rainfall-runoff Transformation, including Water Balance (FEST-WB) and the Weather Research and Forecasting (WRF) models. Accurate forecasts of deep moist convection and extreme precipitation are difficult to be predicted due to uncertainties arising from the numeric weather prediction (NWP) physical parameterizations and high sensitivity to misrepresentation of the atmospheric state; however, two hydrological ensemble prediction systems (HEPS) have been designed to explicitly cope with uncertainties in the initial and lateral boundary conditions (IC/LBCs) and physical parameterizations of the NWP model. No substantial differences in skill have been found between both ensemble strategies when considering an enhanced diversity of IC/LBCs for the perturbed initial conditions ensemble. Furthermore, no additional benefits have been found by considering more frequent LBCs in a mixed physics ensemble, as ensemble spread seems to be reduced. These findings could help to design the most appropriate ensemble strategies before these hydrometeorological extremes, given the computational cost of running such advanced HEPSs for operational purposes.

Rapid sampling of local minima in protein energy surface and effective reduction through a multi-objective filter

PubMed Central

2013-01-01

Background Many problems in protein modeling require obtaining a discrete representation of the protein conformational space as an ensemble of conformations. In ab-initio structure prediction, in particular, where the goal is to predict the native structure of a protein chain given its amino-acid sequence, the ensemble needs to satisfy energetic constraints. Given the thermodynamic hypothesis, an effective ensemble contains low-energy conformations which are similar to the native structure. The high-dimensionality of the conformational space and the ruggedness of the underlying energy surface currently make it very difficult to obtain such an ensemble. Recent studies have proposed that Basin Hopping is a promising probabilistic search framework to obtain a discrete representation of the protein energy surface in terms of local minima. Basin Hopping performs a series of structural perturbations followed by energy minimizations with the goal of hopping between nearby energy minima. This approach has been shown to be effective in obtaining conformations near the native structure for small systems. Recent work by us has extended this framework to larger systems through employment of the molecular fragment replacement technique, resulting in rapid sampling of large ensembles. Methods This paper investigates the algorithmic components in Basin Hopping to both understand and control their effect on the sampling of near-native minima. Realizing that such an ensemble is reduced before further refinement in full ab-initio protocols, we take an additional step and analyze the quality of the ensemble retained by ensemble reduction techniques. We propose a novel multi-objective technique based on the Pareto front to filter the ensemble of sampled local minima. Results and conclusions We show that controlling the magnitude of the perturbation allows directly controlling the distance between consecutively-sampled local minima and, in turn, steering the exploration towards conformations near the native structure. For the minimization step, we show that the addition of Metropolis Monte Carlo-based minimization is no more effective than a simple greedy search. Finally, we show that the size of the ensemble of sampled local minima can be effectively and efficiently reduced by a multi-objective filter to obtain a simpler representation of the probed energy surface. PMID:24564970

Synoptic Factors Affecting Structure Predictability of Hurricane Alex (2016)

NASA Astrophysics Data System (ADS)

Gonzalez-Aleman, J. J.; Evans, J. L.; Kowaleski, A. M.

2016-12-01

On January 7, 2016, a disturbance formed over the western North Atlantic basin. After undergoing tropical transition, the system became the first hurricane of 2016 - and the first North Atlantic hurricane to form in January since 1938. Already an extremely rare hurricane event, Alex then underwent extratropical transition [ET] just north of the Azores Islands. We examine the factors affecting Alex's structural evolution through a new technique called path-clustering. In this way, 51 ensembles from the European Centre for Medium-Range Weather Forecasts Ensemble Prediction System (ECMWF-EPS) are grouped based on similarities in the storm's path through the Cyclone Phase Space (CPS). The differing clusters group various possible scenarios of structural development represented in the ensemble forecasts. As a result, it is possible to shed light on the role of the synoptic scale in changing the structure of this hurricane in the midlatitudes through intercomparison of the most "realistic" forecast of the evolution of Alex and the other physically plausible modes of its development.

Using beta binomials to estimate classification uncertainty for ensemble models.

PubMed

Clark, Robert D; Liang, Wenkel; Lee, Adam C; Lawless, Michael S; Fraczkiewicz, Robert; Waldman, Marvin

2014-01-01

Quantitative structure-activity (QSAR) models have enormous potential for reducing drug discovery and development costs as well as the need for animal testing. Great strides have been made in estimating their overall reliability, but to fully realize that potential, researchers and regulators need to know how confident they can be in individual predictions. Submodels in an ensemble model which have been trained on different subsets of a shared training pool represent multiple samples of the model space, and the degree of agreement among them contains information on the reliability of ensemble predictions. For artificial neural network ensembles (ANNEs) using two different methods for determining ensemble classification - one using vote tallies and the other averaging individual network outputs - we have found that the distribution of predictions across positive vote tallies can be reasonably well-modeled as a beta binomial distribution, as can the distribution of errors. Together, these two distributions can be used to estimate the probability that a given predictive classification will be in error. Large data sets comprised of logP, Ames mutagenicity, and CYP2D6 inhibition data are used to illustrate and validate the method. The distributions of predictions and errors for the training pool accurately predicted the distribution of predictions and errors for large external validation sets, even when the number of positive and negative examples in the training pool were not balanced. Moreover, the likelihood of a given compound being prospectively misclassified as a function of the degree of consensus between networks in the ensemble could in most cases be estimated accurately from the fitted beta binomial distributions for the training pool. Confidence in an individual predictive classification by an ensemble model can be accurately assessed by examining the distributions of predictions and errors as a function of the degree of agreement among the constituent submodels. Further, ensemble uncertainty estimation can often be improved by adjusting the voting or classification threshold based on the parameters of the error distribution. Finally, the profiles for models whose predictive uncertainty estimates are not reliable provide clues to that effect without the need for comparison to an external test set.

A new Method for the Estimation of Initial Condition Uncertainty Structures in Mesoscale Models

NASA Astrophysics Data System (ADS)

Keller, J. D.; Bach, L.; Hense, A.

2012-12-01

The estimation of fast growing error modes of a system is a key interest of ensemble data assimilation when assessing uncertainty in initial conditions. Over the last two decades three methods (and variations of these methods) have evolved for global numerical weather prediction models: ensemble Kalman filter, singular vectors and breeding of growing modes (or now ensemble transform). While the former incorporates a priori model error information and observation error estimates to determine ensemble initial conditions, the latter two techniques directly address the error structures associated with Lyapunov vectors. However, in global models these structures are mainly associated with transient global wave patterns. When assessing initial condition uncertainty in mesoscale limited area models, several problems regarding the aforementioned techniques arise: (a) additional sources of uncertainty on the smaller scales contribute to the error and (b) error structures from the global scale may quickly move through the model domain (depending on the size of the domain). To address the latter problem, perturbation structures from global models are often included in the mesoscale predictions as perturbed boundary conditions. However, the initial perturbations (when used) are often generated with a variant of an ensemble Kalman filter which does not necessarily focus on the large scale error patterns. In the framework of the European regional reanalysis project of the Hans-Ertel-Center for Weather Research we use a mesoscale model with an implemented nudging data assimilation scheme which does not support ensemble data assimilation at all. In preparation of an ensemble-based regional reanalysis and for the estimation of three-dimensional atmospheric covariance structures, we implemented a new method for the assessment of fast growing error modes for mesoscale limited area models. The so-called self-breeding is development based on the breeding of growing modes technique. Initial perturbations are integrated forward for a short time period and then rescaled and added to the initial state again. Iterating this rapid breeding cycle provides estimates for the initial uncertainty structure (or local Lyapunov vectors) given a specific norm. To avoid that all ensemble perturbations converge towards the leading local Lyapunov vector we apply an ensemble transform variant to orthogonalize the perturbations in the sub-space spanned by the ensemble. By choosing different kind of norms to measure perturbation growth, this technique allows for estimating uncertainty patterns targeted at specific sources of errors (e.g. convection, turbulence). With case study experiments we show applications of the self-breeding method for different sources of uncertainty and different horizontal scales.

Prediction of Human Phenotype Ontology terms by means of hierarchical ensemble methods.

PubMed

Notaro, Marco; Schubach, Max; Robinson, Peter N; Valentini, Giorgio

2017-10-12

The prediction of human gene-abnormal phenotype associations is a fundamental step toward the discovery of novel genes associated with human disorders, especially when no genes are known to be associated with a specific disease. In this context the Human Phenotype Ontology (HPO) provides a standard categorization of the abnormalities associated with human diseases. While the problem of the prediction of gene-disease associations has been widely investigated, the related problem of gene-phenotypic feature (i.e., HPO term) associations has been largely overlooked, even if for most human genes no HPO term associations are known and despite the increasing application of the HPO to relevant medical problems. Moreover most of the methods proposed in literature are not able to capture the hierarchical relationships between HPO terms, thus resulting in inconsistent and relatively inaccurate predictions. We present two hierarchical ensemble methods that we formally prove to provide biologically consistent predictions according to the hierarchical structure of the HPO. The modular structure of the proposed methods, that consists in a "flat" learning first step and a hierarchical combination of the predictions in the second step, allows the predictions of virtually any flat learning method to be enhanced. The experimental results show that hierarchical ensemble methods are able to predict novel associations between genes and abnormal phenotypes with results that are competitive with state-of-the-art algorithms and with a significant reduction of the computational complexity. Hierarchical ensembles are efficient computational methods that guarantee biologically meaningful predictions that obey the true path rule, and can be used as a tool to improve and make consistent the HPO terms predictions starting from virtually any flat learning method. The implementation of the proposed methods is available as an R package from the CRAN repository.

NWP model forecast skill optimization via closure parameter variations

NASA Astrophysics Data System (ADS)

Järvinen, H.; Ollinaho, P.; Laine, M.; Solonen, A.; Haario, H.

2012-04-01

We present results of a novel approach to tune predictive skill of numerical weather prediction (NWP) models. These models contain tunable parameters which appear in parameterizations schemes of sub-grid scale physical processes. The current practice is to specify manually the numerical parameter values, based on expert knowledge. We developed recently a concept and method (QJRMS 2011) for on-line estimation of the NWP model parameters via closure parameter variations. The method called EPPES ("Ensemble prediction and parameter estimation system") utilizes ensemble prediction infra-structure for parameter estimation in a very cost-effective way: practically no new computations are introduced. The approach provides an algorithmic decision making tool for model parameter optimization in operational NWP. In EPPES, statistical inference about the NWP model tunable parameters is made by (i) generating an ensemble of predictions so that each member uses different model parameter values, drawn from a proposal distribution, and (ii) feeding-back the relative merits of the parameter values to the proposal distribution, based on evaluation of a suitable likelihood function against verifying observations. In this presentation, the method is first illustrated in low-order numerical tests using a stochastic version of the Lorenz-95 model which effectively emulates the principal features of ensemble prediction systems. The EPPES method correctly detects the unknown and wrongly specified parameters values, and leads to an improved forecast skill. Second, results with an ensemble prediction system emulator, based on the ECHAM5 atmospheric GCM show that the model tuning capability of EPPES scales up to realistic models and ensemble prediction systems. Finally, preliminary results of EPPES in the context of ECMWF forecasting system are presented.

The Impact of Model and Rainfall Forcing Errors on Characterizing Soil Moisture Uncertainty in Land Surface Modeling

NASA Technical Reports Server (NTRS)

Maggioni, V.; Anagnostou, E. N.; Reichle, R. H.

2013-01-01

The contribution of rainfall forcing errors relative to model (structural and parameter) uncertainty in the prediction of soil moisture is investigated by integrating the NASA Catchment Land Surface Model (CLSM), forced with hydro-meteorological data, in the Oklahoma region. Rainfall-forcing uncertainty is introduced using a stochastic error model that generates ensemble rainfall fields from satellite rainfall products. The ensemble satellite rain fields are propagated through CLSM to produce soil moisture ensembles. Errors in CLSM are modeled with two different approaches: either by perturbing model parameters (representing model parameter uncertainty) or by adding randomly generated noise (representing model structure and parameter uncertainty) to the model prognostic variables. Our findings highlight that the method currently used in the NASA GEOS-5 Land Data Assimilation System to perturb CLSM variables poorly describes the uncertainty in the predicted soil moisture, even when combined with rainfall model perturbations. On the other hand, by adding model parameter perturbations to rainfall forcing perturbations, a better characterization of uncertainty in soil moisture simulations is observed. Specifically, an analysis of the rank histograms shows that the most consistent ensemble of soil moisture is obtained by combining rainfall and model parameter perturbations. When rainfall forcing and model prognostic perturbations are added, the rank histogram shows a U-shape at the domain average scale, which corresponds to a lack of variability in the forecast ensemble. The more accurate estimation of the soil moisture prediction uncertainty obtained by combining rainfall and parameter perturbations is encouraging for the application of this approach in ensemble data assimilation systems.

Correlation of chemical shifts predicted by molecular dynamics simulations for partially disordered proteins.

PubMed

Karp, Jerome M; Eryilmaz, Ertan; Erylimaz, Ertan; Cowburn, David

2015-01-01

There has been a longstanding interest in being able to accurately predict NMR chemical shifts from structural data. Recent studies have focused on using molecular dynamics (MD) simulation data as input for improved prediction. Here we examine the accuracy of chemical shift prediction for intein systems, which have regions of intrinsic disorder. We find that using MD simulation data as input for chemical shift prediction does not consistently improve prediction accuracy over use of a static X-ray crystal structure. This appears to result from the complex conformational ensemble of the disordered protein segments. We show that using accelerated molecular dynamics (aMD) simulations improves chemical shift prediction, suggesting that methods which better sample the conformational ensemble like aMD are more appropriate tools for use in chemical shift prediction for proteins with disordered regions. Moreover, our study suggests that data accurately reflecting protein dynamics must be used as input for chemical shift prediction in order to correctly predict chemical shifts in systems with disorder.

Ensemble predictive model for more accurate soil organic carbon spectroscopic estimation

NASA Astrophysics Data System (ADS)

Vašát, Radim; Kodešová, Radka; Borůvka, Luboš

2017-07-01

A myriad of signal pre-processing strategies and multivariate calibration techniques has been explored in attempt to improve the spectroscopic prediction of soil organic carbon (SOC) over the last few decades. Therefore, to come up with a novel, more powerful, and accurate predictive approach to beat the rank becomes a challenging task. However, there may be a way, so that combine several individual predictions into a single final one (according to ensemble learning theory). As this approach performs best when combining in nature different predictive algorithms that are calibrated with structurally different predictor variables, we tested predictors of two different kinds: 1) reflectance values (or transforms) at each wavelength and 2) absorption feature parameters. Consequently we applied four different calibration techniques, two per each type of predictors: a) partial least squares regression and support vector machines for type 1, and b) multiple linear regression and random forest for type 2. The weights to be assigned to individual predictions within the ensemble model (constructed as a weighted average) were determined by an automated procedure that ensured the best solution among all possible was selected. The approach was tested at soil samples taken from surface horizon of four sites differing in the prevailing soil units. By employing the ensemble predictive model the prediction accuracy of SOC improved at all four sites. The coefficient of determination in cross-validation (R2cv) increased from 0.849, 0.611, 0.811 and 0.644 (the best individual predictions) to 0.864, 0.650, 0.824 and 0.698 for Site 1, 2, 3 and 4, respectively. Generally, the ensemble model affected the final prediction so that the maximal deviations of predicted vs. observed values of the individual predictions were reduced, and thus the correlation cloud became thinner as desired.

Quantifying Nucleic Acid Ensembles with X-ray Scattering Interferometry.

PubMed

Shi, Xuesong; Bonilla, Steve; Herschlag, Daniel; Harbury, Pehr

2015-01-01

The conformational ensemble of a macromolecule is the complete description of the macromolecule's solution structures and can reveal important aspects of macromolecular folding, recognition, and function. However, most experimental approaches determine an average or predominant structure, or follow transitions between states that each can only be described by an average structure. Ensembles have been extremely difficult to experimentally characterize. We present the unique advantages and capabilities of a new biophysical technique, X-ray scattering interferometry (XSI), for probing and quantifying structural ensembles. XSI measures the interference of scattered waves from two heavy metal probes attached site specifically to a macromolecule. A Fourier transform of the interference pattern gives the fractional abundance of different probe separations directly representing the multiple conformation states populated by the macromolecule. These probe-probe distance distributions can then be used to define the structural ensemble of the macromolecule. XSI provides accurate, calibrated distance in a model-independent fashion with angstrom scale sensitivity in distances. XSI data can be compared in a straightforward manner to atomic coordinates determined experimentally or predicted by molecular dynamics simulations. We describe the conceptual framework for XSI and provide a detailed protocol for carrying out an XSI experiment. © 2015 Elsevier Inc. All rights reserved.

Protein-protein structure prediction by scoring molecular dynamics trajectories of putative poses.

PubMed

Sarti, Edoardo; Gladich, Ivan; Zamuner, Stefano; Correia, Bruno E; Laio, Alessandro

2016-09-01

The prediction of protein-protein interactions and their structural configuration remains a largely unsolved problem. Most of the algorithms aimed at finding the native conformation of a protein complex starting from the structure of its monomers are based on searching the structure corresponding to the global minimum of a suitable scoring function. However, protein complexes are often highly flexible, with mobile side chains and transient contacts due to thermal fluctuations. Flexibility can be neglected if one aims at finding quickly the approximate structure of the native complex, but may play a role in structure refinement, and in discriminating solutions characterized by similar scores. We here benchmark the capability of some state-of-the-art scoring functions (BACH-SixthSense, PIE/PISA and Rosetta) in discriminating finite-temperature ensembles of structures corresponding to the native state and to non-native configurations. We produce the ensembles by running thousands of molecular dynamics simulations in explicit solvent starting from poses generated by rigid docking and optimized in vacuum. We find that while Rosetta outperformed the other two scoring functions in scoring the structures in vacuum, BACH-SixthSense and PIE/PISA perform better in distinguishing near-native ensembles of structures generated by molecular dynamics in explicit solvent. Proteins 2016; 84:1312-1320. © 2016 Wiley Periodicals, Inc. © 2016 Wiley Periodicals, Inc.

Ensemble learning of inverse probability weights for marginal structural modeling in large observational datasets.

PubMed

Gruber, Susan; Logan, Roger W; Jarrín, Inmaculada; Monge, Susana; Hernán, Miguel A

2015-01-15

Inverse probability weights used to fit marginal structural models are typically estimated using logistic regression. However, a data-adaptive procedure may be able to better exploit information available in measured covariates. By combining predictions from multiple algorithms, ensemble learning offers an alternative to logistic regression modeling to further reduce bias in estimated marginal structural model parameters. We describe the application of two ensemble learning approaches to estimating stabilized weights: super learning (SL), an ensemble machine learning approach that relies on V-fold cross validation, and an ensemble learner (EL) that creates a single partition of the data into training and validation sets. Longitudinal data from two multicenter cohort studies in Spain (CoRIS and CoRIS-MD) were analyzed to estimate the mortality hazard ratio for initiation versus no initiation of combined antiretroviral therapy among HIV positive subjects. Both ensemble approaches produced hazard ratio estimates further away from the null, and with tighter confidence intervals, than logistic regression modeling. Computation time for EL was less than half that of SL. We conclude that ensemble learning using a library of diverse candidate algorithms offers an alternative to parametric modeling of inverse probability weights when fitting marginal structural models. With large datasets, EL provides a rich search over the solution space in less time than SL with comparable results. Copyright © 2014 John Wiley & Sons, Ltd.

Ensemble learning of inverse probability weights for marginal structural modeling in large observational datasets

PubMed Central

Gruber, Susan; Logan, Roger W.; Jarrín, Inmaculada; Monge, Susana; Hernán, Miguel A.

2014-01-01

Inverse probability weights used to fit marginal structural models are typically estimated using logistic regression. However a data-adaptive procedure may be able to better exploit information available in measured covariates. By combining predictions from multiple algorithms, ensemble learning offers an alternative to logistic regression modeling to further reduce bias in estimated marginal structural model parameters. We describe the application of two ensemble learning approaches to estimating stabilized weights: super learning (SL), an ensemble machine learning approach that relies on V -fold cross validation, and an ensemble learner (EL) that creates a single partition of the data into training and validation sets. Longitudinal data from two multicenter cohort studies in Spain (CoRIS and CoRIS-MD) were analyzed to estimate the mortality hazard ratio for initiation versus no initiation of combined antiretroviral therapy among HIV positive subjects. Both ensemble approaches produced hazard ratio estimates further away from the null, and with tighter confidence intervals, than logistic regression modeling. Computation time for EL was less than half that of SL. We conclude that ensemble learning using a library of diverse candidate algorithms offers an alternative to parametric modeling of inverse probability weights when fitting marginal structural models. With large datasets, EL provides a rich search over the solution space in less time than SL with comparable results. PMID:25316152

Ensemble MD simulations restrained via crystallographic data: Accurate structure leads to accurate dynamics

PubMed Central

Xue, Yi; Skrynnikov, Nikolai R

2014-01-01

Currently, the best existing molecular dynamics (MD) force fields cannot accurately reproduce the global free-energy minimum which realizes the experimental protein structure. As a result, long MD trajectories tend to drift away from the starting coordinates (e.g., crystallographic structures). To address this problem, we have devised a new simulation strategy aimed at protein crystals. An MD simulation of protein crystal is essentially an ensemble simulation involving multiple protein molecules in a crystal unit cell (or a block of unit cells). To ensure that average protein coordinates remain correct during the simulation, we introduced crystallography-based restraints into the MD protocol. Because these restraints are aimed at the ensemble-average structure, they have only minimal impact on conformational dynamics of the individual protein molecules. So long as the average structure remains reasonable, the proteins move in a native-like fashion as dictated by the original force field. To validate this approach, we have used the data from solid-state NMR spectroscopy, which is the orthogonal experimental technique uniquely sensitive to protein local dynamics. The new method has been tested on the well-established model protein, ubiquitin. The ensemble-restrained MD simulations produced lower crystallographic R factors than conventional simulations; they also led to more accurate predictions for crystallographic temperature factors, solid-state chemical shifts, and backbone order parameters. The predictions for 15N R1 relaxation rates are at least as accurate as those obtained from conventional simulations. Taken together, these results suggest that the presented trajectories may be among the most realistic protein MD simulations ever reported. In this context, the ensemble restraints based on high-resolution crystallographic data can be viewed as protein-specific empirical corrections to the standard force fields. PMID:24452989

Self-Adaptive Prediction of Cloud Resource Demands Using Ensemble Model and Subtractive-Fuzzy Clustering Based Fuzzy Neural Network

PubMed Central

Chen, Zhijia; Zhu, Yuanchang; Di, Yanqiang; Feng, Shaochong

2015-01-01

In IaaS (infrastructure as a service) cloud environment, users are provisioned with virtual machines (VMs). To allocate resources for users dynamically and effectively, accurate resource demands predicting is essential. For this purpose, this paper proposes a self-adaptive prediction method using ensemble model and subtractive-fuzzy clustering based fuzzy neural network (ESFCFNN). We analyze the characters of user preferences and demands. Then the architecture of the prediction model is constructed. We adopt some base predictors to compose the ensemble model. Then the structure and learning algorithm of fuzzy neural network is researched. To obtain the number of fuzzy rules and the initial value of the premise and consequent parameters, this paper proposes the fuzzy c-means combined with subtractive clustering algorithm, that is, the subtractive-fuzzy clustering. Finally, we adopt different criteria to evaluate the proposed method. The experiment results show that the method is accurate and effective in predicting the resource demands. PMID:25691896

Prediction of protein long-range contacts using an ensemble of genetic algorithm classifiers with sequence profile centers.

PubMed

Chen, Peng; Li, Jinyan

2010-05-17

Prediction of long-range inter-residue contacts is an important topic in bioinformatics research. It is helpful for determining protein structures, understanding protein foldings, and therefore advancing the annotation of protein functions. In this paper, we propose a novel ensemble of genetic algorithm classifiers (GaCs) to address the long-range contact prediction problem. Our method is based on the key idea called sequence profile centers (SPCs). Each SPC is the average sequence profiles of residue pairs belonging to the same contact class or non-contact class. GaCs train on multiple but different pairs of long-range contact data (positive data) and long-range non-contact data (negative data). The negative data sets, having roughly the same sizes as the positive ones, are constructed by random sampling over the original imbalanced negative data. As a result, about 21.5% long-range contacts are correctly predicted. We also found that the ensemble of GaCs indeed makes an accuracy improvement by around 5.6% over the single GaC. Classifiers with the use of sequence profile centers may advance the long-range contact prediction. In line with this approach, key structural features in proteins would be determined with high efficiency and accuracy.

Predicting nucleic acid binding interfaces from structural models of proteins

PubMed Central

Dror, Iris; Shazman, Shula; Mukherjee, Srayanta; Zhang, Yang; Glaser, Fabian; Mandel-Gutfreund, Yael

2011-01-01

The function of DNA- and RNA-binding proteins can be inferred from the characterization and accurate prediction of their binding interfaces. However the main pitfall of various structure-based methods for predicting nucleic acid binding function is that they are all limited to a relatively small number of proteins for which high-resolution three dimensional structures are available. In this study, we developed a pipeline for extracting functional electrostatic patches from surfaces of protein structural models, obtained using the I-TASSER protein structure predictor. The largest positive patches are extracted from the protein surface using the patchfinder algorithm. We show that functional electrostatic patches extracted from an ensemble of structural models highly overlap the patches extracted from high-resolution structures. Furthermore, by testing our pipeline on a set of 55 known nucleic acid binding proteins for which I-TASSER produces high-quality models, we show that the method accurately identifies the nucleic acids binding interface on structural models of proteins. Employing a combined patch approach we show that patches extracted from an ensemble of models better predicts the real nucleic acid binding interfaces compared to patches extracted from independent models. Overall, these results suggest that combining information from a collection of low-resolution structural models could be a valuable approach for functional annotation. We suggest that our method will be further applicable for predicting other functional surfaces of proteins with unknown structure. PMID:22086767

Correlated variability modifies working memory fidelity in primate prefrontal neuronal ensembles

PubMed Central

Leavitt, Matthew L.; Pieper, Florian; Sachs, Adam J.; Martinez-Trujillo, Julio C.

2017-01-01

Neurons in the primate lateral prefrontal cortex (LPFC) encode working memory (WM) representations via sustained firing, a phenomenon hypothesized to arise from recurrent dynamics within ensembles of interconnected neurons. Here, we tested this hypothesis by using microelectrode arrays to examine spike count correlations (rsc) in LPFC neuronal ensembles during a spatial WM task. We found a pattern of pairwise rsc during WM maintenance indicative of stronger coupling between similarly tuned neurons and increased inhibition between dissimilarly tuned neurons. We then used a linear decoder to quantify the effects of the high-dimensional rsc structure on information coding in the neuronal ensembles. We found that the rsc structure could facilitate or impair coding, depending on the size of the ensemble and tuning properties of its constituent neurons. A simple optimization procedure demonstrated that near-maximum decoding performance could be achieved using a relatively small number of neurons. These WM-optimized subensembles were more signal correlation (rsignal)-diverse and anatomically dispersed than predicted by the statistics of the full recorded population of neurons, and they often contained neurons that were poorly WM-selective, yet enhanced coding fidelity by shaping the ensemble’s rsc structure. We observed a pattern of rsc between LPFC neurons indicative of recurrent dynamics as a mechanism for WM-related activity and that the rsc structure can increase the fidelity of WM representations. Thus, WM coding in LPFC neuronal ensembles arises from a complex synergy between single neuron coding properties and multidimensional, ensemble-level phenomena. PMID:28275096

Ensemble Nonlinear Autoregressive Exogenous Artificial Neural Networks for Short-Term Wind Speed and Power Forecasting.

PubMed

Men, Zhongxian; Yee, Eugene; Lien, Fue-Sang; Yang, Zhiling; Liu, Yongqian

2014-01-01

Short-term wind speed and wind power forecasts (for a 72 h period) are obtained using a nonlinear autoregressive exogenous artificial neural network (ANN) methodology which incorporates either numerical weather prediction or high-resolution computational fluid dynamics wind field information as an exogenous input. An ensemble approach is used to combine the predictions from many candidate ANNs in order to provide improved forecasts for wind speed and power, along with the associated uncertainties in these forecasts. More specifically, the ensemble ANN is used to quantify the uncertainties arising from the network weight initialization and from the unknown structure of the ANN. All members forming the ensemble of neural networks were trained using an efficient particle swarm optimization algorithm. The results of the proposed methodology are validated using wind speed and wind power data obtained from an operational wind farm located in Northern China. The assessment demonstrates that this methodology for wind speed and power forecasting generally provides an improvement in predictive skills when compared to the practice of using an "optimal" weight vector from a single ANN while providing additional information in the form of prediction uncertainty bounds.

Ensemble Nonlinear Autoregressive Exogenous Artificial Neural Networks for Short-Term Wind Speed and Power Forecasting

PubMed Central

Lien, Fue-Sang; Yang, Zhiling; Liu, Yongqian

2014-01-01

Short-term wind speed and wind power forecasts (for a 72 h period) are obtained using a nonlinear autoregressive exogenous artificial neural network (ANN) methodology which incorporates either numerical weather prediction or high-resolution computational fluid dynamics wind field information as an exogenous input. An ensemble approach is used to combine the predictions from many candidate ANNs in order to provide improved forecasts for wind speed and power, along with the associated uncertainties in these forecasts. More specifically, the ensemble ANN is used to quantify the uncertainties arising from the network weight initialization and from the unknown structure of the ANN. All members forming the ensemble of neural networks were trained using an efficient particle swarm optimization algorithm. The results of the proposed methodology are validated using wind speed and wind power data obtained from an operational wind farm located in Northern China. The assessment demonstrates that this methodology for wind speed and power forecasting generally provides an improvement in predictive skills when compared to the practice of using an “optimal” weight vector from a single ANN while providing additional information in the form of prediction uncertainty bounds. PMID:27382627

Predication of different stages of Alzheimer's disease using neighborhood component analysis and ensemble decision tree.

PubMed

Jin, Mingwu; Deng, Weishu

2018-05-15

There is a spectrum of the progression from healthy control (HC) to mild cognitive impairment (MCI) without conversion to Alzheimer's disease (AD), to MCI with conversion to AD (cMCI), and to AD. This study aims to predict the different disease stages using brain structural information provided by magnetic resonance imaging (MRI) data. The neighborhood component analysis (NCA) is applied to select most powerful features for prediction. The ensemble decision tree classifier is built to predict which group the subject belongs to. The best features and model parameters are determined by cross validation of the training data. Our results show that 16 out of a total of 429 features were selected by NCA using 240 training subjects, including MMSE score and structural measures in memory-related regions. The boosting tree model with NCA features can achieve prediction accuracy of 56.25% on 160 test subjects. Principal component analysis (PCA) and sequential feature selection (SFS) are used for feature selection, while support vector machine (SVM) is used for classification. The boosting tree model with NCA features outperforms all other combinations of feature selection and classification methods. The results suggest that NCA be a better feature selection strategy than PCA and SFS for the data used in this study. Ensemble tree classifier with boosting is more powerful than SVM to predict the subject group. However, more advanced feature selection and classification methods or additional measures besides structural MRI may be needed to improve the prediction performance. Copyright © 2018 Elsevier B.V. All rights reserved.

Decadal climate predictions improved by ocean ensemble dispersion filtering

NASA Astrophysics Data System (ADS)

Kadow, C.; Illing, S.; Kröner, I.; Ulbrich, U.; Cubasch, U.

2017-06-01

Decadal predictions by Earth system models aim to capture the state and phase of the climate several years in advance. Atmosphere-ocean interaction plays an important role for such climate forecasts. While short-term weather forecasts represent an initial value problem and long-term climate projections represent a boundary condition problem, the decadal climate prediction falls in-between these two time scales. In recent years, more precise initialization techniques of coupled Earth system models and increased ensemble sizes have improved decadal predictions. However, climate models in general start losing the initialized signal and its predictive skill from one forecast year to the next. Here we show that the climate prediction skill of an Earth system model can be improved by a shift of the ocean state toward the ensemble mean of its individual members at seasonal intervals. We found that this procedure, called ensemble dispersion filter, results in more accurate results than the standard decadal prediction. Global mean and regional temperature, precipitation, and winter cyclone predictions show an increased skill up to 5 years ahead. Furthermore, the novel technique outperforms predictions with larger ensembles and higher resolution. Our results demonstrate how decadal climate predictions benefit from ocean ensemble dispersion filtering toward the ensemble mean.