Explicit Statistical Model Research Articles

Drivers of land abandonment in Southern Chile and implications for landscape planning

In recent decades, land abandonment has become one of the most important trends in land use and cover change. This study identifies drivers of land abandonment in Southern Chile through a spatially explicit statistical model, based on an economic theoretical framework. A logistic regression was used to model the probability of land abandonment, observed as the land cover change from agricultural land to arboreous shrubland or native forest between 1985 and 2007. We relied on census information and land cover data from 2472 farms, which represented 86.6% of the farm population in the study area. Land cover data came from two Landsat scenes taken in 1985 (TM) and 2007 (ETM+). The results from the spatial analysis show that agricultural land covered 30 061 ha in 1985. By 2007, only 16 518 ha had persisted in agriculture, while the rest had been covered by natural vegetation. Soil quality was a significant benefit-related geophysical driver of land abandonment. Important cost related socioeconomic drivers were the distances to secondary roads, aquaculture production centers, and national parks, and the existence of farm subsidies. Significant farm structural variables were the amount of bovine heads and farm's livestock carrying capacity. Variables such as age, education, and place of residence of the farmer were not statistically significant. These results may serve as an important planning tool for land managers, by helping identify areas and farming systems with high probability of land abandonment, and furthermore explore the outcomes of some potential scenarios on the future probability of the process.

Models and Methods for Analyzing Phenotypic Evolution in Lineages and Clades

In this chapter we discuss methods for analyzing continuous traits, with an emphasis on those approaches that rely on explicit statistical models of evolution and incorporate genealogical information (ancestor–descendant or phylogenetic relationships). After discussing the roles of models and genealogy in evolutionary inference, we summarize the properties of commonly used models including random walks (Brownian motion), directional evolution, and stasis. These models can be used to devise null-hypothesis tests about evolutionary patterns, but it is often better to fit and compare models equally using information criteria and related approaches. We apply these methods to a published data set of dental measurements in a sequence of ancestor–descendant populations in the early primateCantius, with the particular goal of determining the best-supported mode of evolutionary change in this lineage. We also assess a series of questions about the evolution of femoral dimensions in several clades of dinosaurs, including testing for a trend of increasing body size (Cope's Rule), testing for correlations among characters, and reconstructing ancestral states. Finally, we list briefly some additional models, approaches, and issues that arise in genealogically informed analyses of phenotypic evolution.

Prediction of alternative isoforms from exon expression levels in RNA-Seq experiments

Alternative splicing, polyadenylation of pre-messenger RNA molecules and differential promoter usage can produce a variety of transcript isoforms whose respective expression levels are regulated in time and space, thus contributing specific biological functions. However, the repertoire of mammalian alternative transcripts and their regulation are still poorly understood. Second-generation sequencing is now opening unprecedented routes to address the analysis of entire transcriptomes. Here, we developed methods that allow the prediction and quantification of alternative isoforms derived solely from exon expression levels in RNA-Seq data. These are based on an explicit statistical model and enable the prediction of alternative isoforms within or between conditions using any known gene annotation, as well as the relative quantification of known transcript structures. Applying these methods to a human RNA-Seq dataset, we validated a significant fraction of the predictions by RT-PCR. Data further showed that these predictions correlated well with information originating from junction reads. A direct comparison with exon arrays indicated improved performances of RNA-Seq over microarrays in the prediction of skipped exons. Altogether, the set of methods presented here comprehensively addresses multiple aspects of alternative isoform analysis. The software is available as an open-source R-package called Solas at http://cmb.molgen.mpg.de/2ndGenerationSequencing/Solas/.

P300 Does in vitro antimicrobial resistance predict clinical outcomes in urinary tract infections: systematic review and metaanalysis

The breakdown of socialism caused massive socio-economic and institutional changes that led to substantial agricultural land abandonment. The goal of our study was to identify the determinants of agricultural land abandonment in post-Soviet Russia during the first decade of transition from a state-controlled economy to a market-driven economy (1990–2000). We analyzed the determinants of agricultural land abandonment for approximately 150,550 km2 of land area in the provinces (oblasts) of Kaluga, Rjazan, Smolensk, Tula and Vladimir in European Russia. Based on the economic assumptions of profit maximization, we integrated maps of abandoned agricultural land from five ∼185 km × 185 km Landsat TM/ETM+ footprints with socio-economic, environmental and geographic variables, and we estimated logistic regressions at the pixel level to identify the determinants of agricultural land abandonment. Our results showed that a higher likelihood of agricultural land abandonment was significantly associated with lower average grain yields in the late 1980s and with higher distances from the nearest settlements, municipality centers, and settlements with more than 500 citizens. Hierarchical partitioning showed that the average grain yields in the late 1980s had the greatest power to explain agricultural land abandonment in our models, followed by the locational attributes of the agricultural land. We hypothesize that the termination of 90% of state subsidies for agriculture from 1990 to 2000 was an important underlying cause for the decrease of cultivation in economically and environmentally marginal agriculture areas. Thus, whereas the spatial patterns corresponded to the land rent theory of von Thünen, it was primarily the macro-scale driving forces that fostered agricultural abandonment. Our study highlighted the value of spatially explicit statistical models for studying the determinants of land-use and land-cover change in large areas.

Fast Statistical Alignment

We describe a new program for the alignment of multiple biological sequences that is both statistically motivated and fast enough for problem sizes that arise in practice. Our Fast Statistical Alignment program is based on pair hidden Markov models which approximate an insertion/deletion process on a tree and uses a sequence annealing algorithm to combine the posterior probabilities estimated from these models into a multiple alignment. FSA uses its explicit statistical model to produce multiple alignments which are accompanied by estimates of the alignment accuracy and uncertainty for every column and character of the alignment—previously available only with alignment programs which use computationally-expensive Markov Chain Monte Carlo approaches—yet can align thousands of long sequences. Moreover, FSA utilizes an unsupervised query-specific learning procedure for parameter estimation which leads to improved accuracy on benchmark reference alignments in comparison to existing programs. The centroid alignment approach taken by FSA, in combination with its learning procedure, drastically reduces the amount of false-positive alignment on biological data in comparison to that given by other methods. The FSA program and a companion visualization tool for exploring uncertainty in alignments can be used via a web interface at http://orangutan.math.berkeley.edu/fsa/, and the source code is available at http://fsa.sourceforge.net/.

Assessing land-use statistics to model land cover change in a mountainous landscape in the European Alps

One of the predominant processes of land cover change in the European Alps over the last 150 years has been the abandonment of agricultural land and the subsequent regeneration of forest. Here, we employed two sequential datasets from Switzerland (for the periods 1979–1985 and 1992–1997) to show how land-use and land cover data can be used to investigate such large scale ecological and land cover change processes. We applied a combination of generalized additive and generalized linear modelling to develop spatially explicit statistical models for land cover transitions between any of the following types: intensively used agricultural land, extensively used agricultural land, overgrown areas, open canopy forest, closed canopy forest. Climate, soil, relief-related data, basic socio-economic variables and information about the composition of the surrounding neighbourhood of the samples were utilised as potential predictors of land cover change. The proportion of variance explained differed considerably between models but a consistently high AUC for both calibration and evaluation datasets was achieved for the majority of the 25, with resulting values ranging from 0.58 to 0.96. The model residuals showed some degree of spatial autocorrelation despite the use of a sparse sampling regime and the inclusion of neighbourhood variables. We conclude that the analysis of sequential land cover datasets using the kind of statistical models developed here offers a promising way to investigate ecological processes such as forest succession at a large spatial scale.

Gradual or pulsed evolution: when should punctuational explanations be preferred?

The problem of gradual versus punctuated change within phyletic lineages can be understood in terms of the homogeneity of evolutionary dynamics. Hypotheses of punctuated change imply that the rules governing evolutionary change shift over time such that the normal dynamics of stasis are temporarily suspended, permitting a period of net evolutionary change. Such explanations are members of a larger class of models in which evolutionary dynamics are in some way heterogeneous over time. In this paper, I develop a likelihood-based statistical framework to evaluate the support for this kind of evolutionary model. This approach divides evolutionary sequences into nonoverlapping segments, each of which is fit to a separate evolutionary model. Models with heterogeneous dynamics are generally more complex—they require more parameters to specify—than uniform evolutionary models such as random walks and stasis. The Akaike Information Criterion can be used to judge whether the greater complexity of punctuational models is offset by a sufficient gain in log-likelihood for these models to be preferred.I use this approach to analyze three case studies for which punctuational explanations have been proposed. In the first, a model of punctuated evolution best accounted for changes in pygidial morphology within a lineage of the trilobiteFlexicalymene, but the uniform model of an unbiased random walk remains a plausible alternative. Body size evolution in the radiolarianPseudocubus vemawas neither purely gradual nor completely pulsed. Instead, the best-supported explanation posited a single, pulsed increase, followed later by a shift to an unbiased random walk. Finally, for the much-analyzed claim of “punctuated gradualism“ in the foraminiferaGloborotalia, the best-supported model implied two periods of stasis separated by a period of elevated but not inherently directional evolution. Although the conclusions supported by these analyses generally refined rather than overturned previous views, the present approach differs from those prior in that all competing interpretations were formalized into explicit statistical models, allowing their relative support to be unambiguously compared.

Statistical Analysis of Visual Fits: Answer to J. Ninio

REPLY: We thank Dr. Ninio for his interest in our 25-yr-old publications and for his attempt to analyze our data and interpretations using a strategy different from the one we used for this pioneering work. However, his approach is exclusively concerned with curve fitting. It avoids the more demanding problem of the recovery of parameters of explicit statistical models of biological processes. That is, he decided to discriminate between fits of our published histograms visually, or “by eye.” The field long ago progressed to the point where the standard for confronting data with models is statistical analysis, not visual fitting. Reevaluation of old methods can be a good thing (Lowen et al. 1997), provided there is a sound scientific questions at stake. Yet, we are surprised that Dr. Ninio raises concerns about the validity of aspects of these seminal papers but does not want to review the subsequent literature that appeared about our results and conclusions. This is difficult to understand because the aim of a meaningful scientific approach is to progress in the understanding of natural phenomena and processes. In this respect, it is worth noting that subsequent research carried out independently by a number of eminent scientists supported our proposal of the “One Vesicle Hypothesis” of transmitter release at several synapses in the CNS. In this response we address the flaws in the logic of Ninio’s analysis and his criticisms and misunderstandings of our published work. Because of the progress in this field, and given that Dr. Ninio is not an expert in this area, we believe our response here will be of interest to other investigators who are concerned with using rigorous methods to extract information from experimental distributions. BACKGROUND Experimental context We performed simultaneous intracellular recordings and dye injections, in vivo, to collect data sets of fluctuating inhibitory postsynaptic potentials (IPSPs) evoked in the Mauthner cell of teleost fish by single presynaptic action potentials and used an optimization procedure to determine the best-fitting binomial and Poisson descriptions of the data. We then compared those results with histological reconstructions of the presynaptic cells, including the axonal branching pattern and the identification of contacts with the postsynaptic neuron. This was one of the earliest attempts to obtain such structure‐function correlations. We were particularly interested in two questions that had been raised previously: 1) Were the fluctuations in PSP amplitude better described by a binomial or a Poisson distribution? and 2) Could the statistical parameters be correlated with physical structures?

TUGAI: An Integrated Simulation Tool for Ecological Assessment of Alternative Water Management Strategies in a Degraded River Delta

The development of ecologically sound water allocation strategies that account for the needs of riverine ecosystems is a pressing issue, especially in semiarid river basins. In the Aral Sea Basin, a search for strategies to mitigate ecological and socioeconomic deterioration has been in process since the early 1990s. The Geographic Information System-based simulation tool TUGAI has been developed to support the policy determination process by providing a simple, problem-oriented method to assess ecological effects of alternative water management strategies for the Amudarya River. It combines a multiobjective water allocation model with simple, spatially explicit statistical and rule-based models of landscape dynamics. Changes in environmental conditions are evaluated by a fuzzy habitat suitability index for Populus euphratica, which is the dominant species of the characteristic riverine Tugai forests. Water management scenarios can be developed by altering spatiotemporal water distribution in the delta area or the amount of water inflow into the delta. Outcomes of scenario analysis are qualitative comparisons of the ecological effects of different options for a time period of up to 28 years. The given approach utilizes different types of knowledge, from quantitative hydrological data to qualitative local expert knowledge. The main purpose of the tool is to integrate the knowledge in a comprehensive way to make it available for discussions on alternative policies in moderated workshops with stakeholders. In this article, the modules of the tool, their integration, and three hypothetical scenarios are presented. Based on the experience gained when developing the TUGAI tool, we propose that the general framework can be transferred to other areas where tradeoffs in water allocation between the environment and other water users are of major concern. The potential for a simulation tool to structure and inform a complex resource management situation by involving local experts and stakeholders in the development of possible future scenarios will become increasingly valuable for transparent and participatory resource management.

A Multimetric Assessment of Stream Condition in the Northern Lakes and Forests Ecoregion Using Spatially Explicit Statistical Modeling and Regional Normalization

AbstractWe sampled fish communities, water temperature, water chemistry, physical habitat, and catchment characteristics for 94 stream sites selected randomly throughout the Northern Lakes and Forests ecoregion and used those data to explicitly model reference conditions and assess ecological stream condition at each site via a regional normalization framework. The streams we sampled were first order through fourth order, and the catchments ranged from 0.9 to 458 km2. We developed multiple linear regression (MLR) models that predicted fish community metrics, water chemistry characteristics, and local physical habitat from catchment characteristics; we used these models to compare existing conditions with the conditions that would be expected based on the regression models. Our results indicated that the fish communities were relatively unimpaired because the catchment variables associated with human‐induced land use change were important in only 1 of the 10 fish metric models. Agricultural land use was a significant variable in the MLR equation for species of Lepomis (sunfish). Agricultural land use and urban land use were both significant variables in all of the MLR models predicting water chemistry variables; urban land use was a significant variable in the MLR model predicting the percent coverage of all instream cover types. Regional normalization indicated that none of the sites were impaired based on fish community attributes. However, our analysis based on water chemistry metrics indicated that 22– 35% of the sites were impaired and that, based on physical habitat, 6–14% of the sites were impaired. A comparison with other published studies of the ecoregion suggested that the regional normalization process correctly characterized stream condition.

An Approach to Proper Speech Segmentation for Quality Improvement in Concatenative Text-To-Speech System for Indian Languages

Most of the Indian-language Text-To-Speech (TTS) synthesis systems designed till date are based upon the concatenation of acoustic units. The prime challenge is the selection of proper units and their elegant concatenation. Due to the precincts of current automated techniques based on Hidden Markov Model (HMM) and Dynamic Time Warping (DTW), manual verification and labeling are often essential. Automatic placement of phoneme boundaries in a speech waveform using explicit statistical model for phoneme boundary is proposed in this paper. We are projecting the Harmonic plus Noise Model (HNM) in the first step and refine the boundary placement by searching for the best match in a region near the estimated boundary with predefined boundary model Technique like ESNOLA. This technique is applied for effective concatenation, which results in smooth output. Studies show that HNM is capable of synthesizing all vowels and diphones with good quality. This can remarkably reduce the size of the database. Further the pitch synchronous analysis and the Glottal Closure Instants (GCI) are accurately calculated. The quality of the synthesized speech improves if these units are obtained from the glottal signal rather than from processing the signal. The database has to be developed for VCV for all Indian languages as we have done for Oriya, one of the official languages of the Republic of India for our case study.

Exploratory Data Analysis for Complex Models

“Exploratory” and “confirmatory” data analysis can both be viewed as methods for comparing observed data to what would be obtained under an implicit or explicit statistical model. For example, many of Tukey's methods can be interpreted as checks against hypothetical linear models and Poisson distributions. In more complex situations, Bayesian methods can be useful for constructing reference distributions for various plots that are useful in exploratory data analysis. This article proposes an approach to unify exploratory data analysis with more formal statistical methods based on probability models. These ideas are developed in the context of examples from fields including psychology, medicine, and social science.

Computational Discovery of Gene Regulatory Binding Motifs: A Bayesian Perspective

The Bayesian approach together with Markov chain Monte Carlo techniques has provided an attractive solution to many important bioinformatics problems such as multiple sequence alignment, microarray analysis and the discovery of gene regulatory binding motifs. The employment of such methods and, more broadly, explicit statistical modeling, has revolutionized the field of computational biology. After reviewing several heuristics-based computational methods, this article presents a systematic account of Bayesian formulations and solutions to the motif discovery problem. Generalizations are made to further enhance the Bayesian approach. Motivated by the need of a speedy algorithm, we also provide a perspective of the problem from the viewpoint of optimizing a scoring function. We observe that scoring functions resulting from proper posterior distributions, or approximations to such distributions, showed the best performance and can be used to improve upon existing motif-finding programs. Simulation analyses and a real-data example are used to support our observation.

On the interpretation and analysis of a degree-of-equivalence

The degree-of-equivalence between the standards realized by two laboratories comprises the difference between these standards and the uncertainty in this difference. However, the exact way in which these two pieces of information are to be used has not been prescribed. This paper shows how the degree-of-equivalence can be used for three different purposes: (i) to assess the mutual compatibility of the standards with a view to identifying problems in one or other measurement procedure; (ii) to decide if the standards are interchangeable without modification for a specific application; and (iii) to obtain a suitable adjustment term and its uncertainty when one standard is to be used with modification in place of another. In particular, the degree-of-equivalence is used with an explicit statistical model to define a `degree-of-compatibility' for the first purpose and a `degree-of-interchangeability' for the second purpose. The analysis is also applicable for the degree-of-equivalence between a participating laboratory and a comparison reference value.

Landscape characteristics, land use, and coho salmon (Oncorhynchus kisutch) abundance, Snohomish River, Wash., U.S.A.

We used temporally consistent patterns in the spatial distribution of returning adult coho salmon (Oncorhynchus kisutch) to explore relationships between salmon abundance, landscape characteristics, and land use patterns in the Snohomish River watershed, Wash. The proportion of total adult coho salmon abundance supported by a specific stream reach was consistent among years, even though interannual adult coho salmon abundance varied substantially. Wetland occurrence, local geology, stream gradient, and land use were significantly correlated with adult coho salmon abundance. Median adult coho salmon densities in forest-dominated areas were 1.5–3.5 times the densities in rural, urban, and agricultural areas. Relationships between these habitat characteristics and adult coho salmon abundance were consistent over time. Spatially explicit statistical models that included these habitat variables explained almost half of the variation in the annual distribution of adult coho salmon. Our analysis indicates that such models can be used to identify and prioritize freshwater areas for protection and restoration.

Algorithms for suppressing ultrasonic backscattering from material structure

In pulse-echo ultrasonic inspection the backscattering from the material structure appears in the received ultrasonic images as clutter, often referred to as grain noise, which impairs the inspection results. A toolbox including algorithms for suppressing ultrasonic clutter is presented in the paper. Several processing algorithms capable of suppressing grain noise have been proposed, of which the split spectrum processing (SSP) probably is the most renowned. The classical SSP technique applies a filter bank to some frequency band that has to be precisely known in advance, to obtain a set of narrow-band signals that are tested for mutual correlation using some statistical operation. A number of SSP algorithms with different statistical operations are included in the toolbox. A completely different approach is to use explicit statistical models of grain noise and defects and to design an optimal filter based on those models. A simple such algorithm, based on noncoherent detection (NCD) known from communications, is also included in the toolbox. The toolbox, implemented in Matlab™, is provided with a user-friendly graphical interface facilitating comparison of the algorithms.

Modelling near‐infrared signals for on‐line monitoring in cheese manufacture

AbstractThis paper considers the analysis of a continuously monitored near‐infrared reflectance signal at a single wavelength for the calibration of a process parameter in an application to food engineering for quality control. We describe how we may summarize the information in the observed signals by postulating an explicit statistical model on the signal. An exploratory data analysis may then be carried out on the profile summaries to evaluate whether and how the functional data provide information on the parameter which we would like to calibrate. From a conceptual point of view, such an approach is not dissimilar to principal component regression methods which use an intermediate decomposition through which information is summarised for calibration of a response. The paper demonstrates how the approach leads to important insights into the manner in which the functional data provide the required information on the desired outcome variable, in the context of the practical application in quality control which is discussed and by using a designed experiment. Calculations are implemented through the Gibbs sampler. Calibration of the prediction equation takes place through meta‐analysis of the summarized profile data in order to take the uncertainty inherent in the summaries into account. Copyright © 2002 John Wiley & Sons, Ltd.

Enhanced speech recognition using an articulatory production model trained on X-ray data

This paper describes an articulatory speech production model trained on an X-ray microbeam database, and presents results of using the model within a speech recognition framework. The system uses an explicit statistical model of co-articulation to increase the accuracy of articulator trajectories synthesized from time-aligned phonetic strings, as compared with X-ray traces. From these trajectories, spectral vector probability distributions are generated using a set of artificial neural networks. The production model is then used in combination with a hidden Markov model recognition system to re-scoreN -best utterance transcription lists. Relative reductions in the word error rate of between 11% and 18% are achieved on a small recognition task.

Information-theoretic analysis of interscale and intrascale dependencies between image wavelet coefficients

This paper presents an information-theoretic analysis of statistical dependencies between image wavelet coefficients. The dependencies are measured using mutual information, which has a fundamental relationship to data compression, estimation, and classification performance. Mutual information is computed analytically for several statistical image models, and depends strongly on the choice of wavelet filters. In the absence of an explicit statistical model, a method is studied for reliably estimating mutual information from image data. The validity of the model-based and data-driven approaches is assessed on representative real-world photographic images. Our results are consistent with empirical observations that coding schemes exploiting inter- and intrascale dependencies alone perform very well, whereas taking both into account does not significantly improve coding performance. A similar observation applies to other image processing applications.

Remarks on the 'Bayesian' method of moments

Zellner has proposed a novel methodology for estimating structural parameters and predicting future observables based on two moments of a subjective distribution and the application of the maximum entropy principle-all in the absence of an explicit statistical model or likelihood function for the data. He calls his procedure the 'Bayesian method of moments' (BMOM). In a recent paper in this journal, Green and Strawderman applied the BMOM to a model for slash pine plantations. It is our view that there are inconsistencies between BMOM and Bayesian (conditional) probability, as we explain in this paper.