Maximum Likelihood-based Methods Research Articles

Asbestos Fiber Concentrations in the Lungs of Brake Repair Workers: An Updated Analysis Using Several Regression Methods to Handle Nondetectable Measurements.

The aim of the study was to reanalyze an updated database of lung asbestos fiber levels for 21 brake repair workers who died of mesothelioma using robust maximum likelihood-based regression methods to address nondetectable measurements. We applied bivariate normal regression to address the doubly left-censored situation where both the lung fiber concentration of noncommercial (TAA) and commercial amphiboles (AC) were subject to detection limits. For the single left-censored situation, we applied censored normal regression to study the relationship between duration of employment (DOE) and TAA. We found a statistically significant positive relationship between TAA and AC (β = 0.49, 95% confidence interval [CI], 0.11 to 0.86) and a not statistically significant relationship between DOE and TAA (β = 0.02, 95% CI, -0.03 to 0.06). Our results provide additional support for the conclusion that exposure to commercial amphibole asbestos, and not chrysotile, is related to the occurrence of mesothelioma among some brake workers.

Maximum Likelihood-Based Methods for Target Velocity Estimation with Distributed MIMO Radar

The estimation problem for target velocity is addressed in this in the scenario with a distributed multi-input multi-out (MIMO) radar system. A maximum likelihood (ML)-based estimation method is derived with the knowledge of target position. Then, in the scenario without the knowledge of target position, an iterative method is proposed to estimate the target velocity by updating the position information iteratively. Moreover, the Carmér-Rao Lower Bounds (CRLBs) for both scenarios are derived, and the performance degradation of velocity estimation without the position information is also expressed. Simulation results show that the proposed estimation methods can approach the CRLBs, and the velocity estimation performance can be further improved by increasing either the number of radar antennas or the information accuracy of the target position. Furthermore, compared with the existing methods, a better estimation performance can be achieved.

Bartlett-type corrections and bootstrap adjustments of likelihood-based inference methods for network meta-analysis.

In network meta-analyses that synthesize direct and indirect comparison evidence concerning multiple treatments, multivariate random effects models have been routinely used for addressing between-studies heterogeneities. Although their standard inference methods depend on large sample approximations (eg, restricted maximum likelihood estimation) for the number of trials synthesized, the numbers of trials are often moderate or small. In these situations, standard estimators cannot be expected to behave in accordance with asymptotic theory; in particular, confidence intervals cannot be assumed to exhibit their nominal coverage probabilities (also, the type I error probabilities of the corresponding tests cannot be retained). The invalidity issue may seriously influence the overall conclusions of network meta-analyses. In this article, we develop several improved inference methods for network meta-analyses to resolve these problems. We first introduce 2 efficient likelihood-based inference methods, the likelihood ratio test-based and efficient score test-based methods, in a general framework of network meta-analysis. Then, to improve the small-sample inferences, we developed improved higher-order asymptotic methods using Bartlett-type corrections and bootstrap adjustment methods. The proposed methods adopt Monte Carlo approaches using parametric bootstraps to effectively circumvent complicated analytical calculations of case-by-case analyses and to permit flexible application to various statistical models network meta-analyses. These methods can also be straightforwardly applied to multivariate meta-regression analyses and to tests for the evaluation of inconsistency. In numerical evaluations via simulations, the proposed methods generally performed well compared with the ordinary restricted maximum likelihood-based inference method. Applications to 2 network meta-analysis datasets are provided.

Compact MQDF classifiers using sparse coding for handwritten Chinese character recognition

The modified quadratic discriminant function (MQDF) is an effective classifier for handwritten Chinese character recognition (HCCR). However, it suffers from high memory requirement for the storage of its parameters, which makes it impractical to be embedded in memory limited hand-held devices. In this paper, we explore the applicability of sparse coding to build compact MQDF classifiers. To be specific, we use sparse coding to compact the parameters of MQDF. Two methods of sparse coding, viz., the maximum likelihood-based method and the K-SVD method, are adopted to build two compact MQDF classifiers, namely, MQDF-ML classifier and MQDF-KSVD classifier. Furthermore, we learn multiple dictionaries rather than single dictionary for sparse coding, because the multiple dictionary learning is capable of not only greatly reducing the computational complexity, but also alleviating the degradation of recognition accuracy, compared to the single dictionary learning. Experiments and comparison with the existing method have demonstrated the effectiveness of our proposed method for the issue of unconstrained handwritten Chinese character recognition.

Stratified UWHAM and Its Stochastic Approximation for Multicanonical Simulations Which Are Far from Equilibrium.

We describe a new analysis tool called Stratified unbinned Weighted Histogram Analysis Method (Stratified-UWHAM), which can be used to compute free energies and expectations from a multicanonical ensemble when a subset of the parallel simulations is far from being equilibrated because of barriers between free energy basins which are only rarely (or never) crossed at some states. The Stratified-UWHAM equations can be obtained in the form of UWHAM equations but with an expanded set of states. We also provide a stochastic solver, Stratified RE-SWHAM, for Stratified-UWHAM to remove its computational bottleneck. Stratified-UWHAM and Stratified RE-SWHAM are applied to study three test topics: the free energy landscape of alanine dipeptide, the binding affinity of a host-guest binding complex, and path sampling for a two-dimensional double well potential. The examples show that when some of the parallel simulations are only locally equilibrated, the estimates of free energies and equilibrium distributions provided by the conventional UWHAM (or MBAR) solutions exhibit considerable biases, but the estimates provided by Stratified-UWHAM and Stratified RE-SWHAM agree with the benchmark very well. Lastly, we discuss features of the Stratified-UWHAM approach which is based on coarse-graining in relation to two other maximum likelihood-based methods which were proposed recently, that also coarse-grain the multicanonical data.

Hallux Valgus, By Nature or Nurture? A Twin Study.

To evaluate the contributions of shared but unmeasured genetic and environmental factors to hallux valgus (HV). Between 2011 and 2012, 74 monozygotic (MZ) and 56 dizygotic (DZ) female twin pairs self-reported HV and putative risk factors, including footwear use across their lifespan. Estimates of casewise concordance (PC ), correlation (ρ), and odds ratios (ORs) were calculated, adjusting for age and other risk factors, and compared between MZ and DZ pairs using logistic regression, generalized estimating equations, and a maximum likelihood-based method, respectively. A total of 70 participants (27%) reported HV, with 12 MZ and 7 DZ pairs being concordant. After adjusting for age, twins were correlated (ρ = 0.27 [95% confidence interval (95% CI) 0.08, 0.46]) and concordant (PC = 0.45 [95% CI 0.29, 0.61]; mean age 58 years), with no difference between MZ and DZ pairs (P = 0.7). HV was associated with regularly wearing footwear with a constrictive toe-box during the fourth decade (adjusted OR 2.73 [95% CI 1.12, 6.67]). This risk factor was correlated in MZ (ρ = 0.38 [95% CI 0.15, 0.60]) but not DZ (ρ = -0.20 [95% CI -0.43, 0.03]) pairs. These correlations were significantly different (P = 0.002). Twins are correlated for HV, but we found no evidence that correlation was due to shared genetic factors. We identified an environmental risk factor, footwear with a constrictive toe-box, that is not shared to the same extent by MZ and DZ pairs, contrary to the assumption of the classic twin model. Footwear, and possibly genetic factors and unknown shared environmental factors, could contribute to developing HV.

Speaker Adaptation of SR-HPM for Speaking Rate-Controlled Mandarin TTS

In this paper, a structural maximum a posteriori (SMAP) speaker adaptation approach to adjusting the speaking rate (SR)-dependent hierarchical prosodic model (SR-HPM) of an existing SR-controlled Mandarin text-to-speech system to a new speaker's data for producing a new voice is discussed. Two main issues are addressed. One is the small SR coverage of the adaptation data and is solved by using the existing SR-HPM that was trained from a speech corpus of wide SR coverage as an informative prior. Another is the data sparseness problem resulting from the large number of parameters of the SR-HPM to be adjusted. It is solved by hierarchically organizing the SR-HPM parameters into decision trees so as to be efficiently adjusted by the SMAP method. The effectiveness of the proposed approach is evaluated on speech databases of five new speakers. Both objective and subjective evaluations show that the proposed method not only performs better than the maximum likelihood-based method in the observed SR range of the target speaker's data, but also is much better in the unseen SR ranges.

Discussion on Maximum Likelihood-Based Methods for Inertial Sensor Calibration

This letter highlights some issues which were overlooked in a recently published paper called maximum likelihood identification of inertial sensor noise model parameters. The latter paper does not consider existing alternative methods, which specifically tackle this issue in a possibly more direct manner and, although remaining a generally valid proposal, does not appear to improve on the earlier proposals. Finally, a simulation study rectifies the poor results of an estimator of reference in the same publication.

Long‐term selection using a single trait criterion, non‐destructive deformation, in White Leghorns: Effect over time on genetic parameters for traits related to egg production

Although non-destructive deformation is relevant for assessing eggshell strength, few long-term selection experiments are documented which use non-destructive deformation as a selection criterion. This study used restricted maximum likelihood-based methods with a four-trait animal model to analyze the effect of non-destructive deformation on egg production, egg weight and sexual maturity in a two-way selection experiment involving 17 generations of White Leghorns. In the strong shell line, corresponding to the line selected for low non-destructive deformation values, the heritability estimates were 0.496 for non-destructive deformation, 0.253 for egg production, 0.660 for egg weight and 0.446 for sexual maturity. In the weak shell line, corresponding to the line selected for high non-destructive deformation values, the heritabilities were 0.372, 0.162, 0.703 and 0.404, respectively. An asymmetric response to selection was observed for non-destructive deformation, egg production and sexual maturity, whereas egg weight decreased for both lines. Using non-destructive deformation to select for stronger eggshell had a small negative effect on egg production and sexual maturity, suggesting the need for breeding programs to balance selection between eggshell traits and egg production traits. However, the analysis of the genetic correlation between non-destructive deformation and egg weight revealed that large eggs are not associated with poor eggshell quality.

Assessment of cardiovascular disease risk factors in a genetically homogenous population of Parsi Zoroastrians in the United States: A pilot study.

Genetically isolated and homogenous populations are ideal for detecting genes underlying common complex diseases. The use of isolated populations with reduced disease heterogeneity has led to significant gene discoveries in the past. The aim of this pilot study was to assess the prevalence of cardiovascular disease (CVD) risk phenotypes in a genetically homogenous population of Parsi Zoroastrians in the United States. Anthropometrics, blood pressure, and medical history were collected from 152 men and 186 women participating in a pilot study as part of the Parsi Family Study. The relative pairs used in the study included 60 parent-off springs, 28 siblings, 6 grandparent-grandchild, 7 avuncular, 18 half-siblings, 7 half-avuncular, and one half-first cousin. Estimates of genetic and environmental influence were calculated using a maximum likelihood-based variance components method implemented in SOLAR. The prevalence of overweight/obesity in adults (62%) was on par with current US prevalence. Hypertension and prehypertension were prevalent in 16% and 46% of the participants, respectively. The quantitative genetic analysis revealed significant heritabilities for all anthropometric phenotypes (P < 0.05). Significant phenotypic correlations were found between blood pressure and anthropometric phenotypes (P < 0.001), whereas significant genetic correlation was found for only diastolic blood pressure and fat free mass (rhoG = -0.88, P < 0.05). These preliminary data show significant additive genetic effects on CVD-related phenotypes in this population. Our findings represent the first epidemiological data in Parsi Zoroastrians in the United States and offer excellent promise for future genetic studies in this population. Am. J. Hum. Biol. 28:440-443, 2016. © 2016 Wiley Periodicals, Inc.

The bimodal distribution of genic GC content is ancestral to monocot species.

In grasses such as rice or maize, the distribution of genic GC content is well known to be bimodal. It is mainly driven by GC content at third codon positions (GC3 for short). This feature is thought to be specific to grasses as closely related species like banana have a unimodal GC3 distribution. GC3 is associated with numerous genomics features and uncovering the origin of this peculiar distribution will help understanding the potential roles and consequences of GC3 variations within and between genomes. Until recently, the origin of the peculiar GC3 distribution in grasses has remained unknown. Thanks to the recent publication of several complete genomes and transcriptomes of nongrass monocots, we studied more than 1,000 groups of one-to-one orthologous genes in seven grasses and three outgroup species (banana, palm tree, and yam). Using a maximum likelihood-based method, we reconstructed GC3 at several ancestral nodes. We found that the bimodal GC3 distribution observed in extant grasses is ancestral to both grasses and most monocot species, and that other species studied here have lost this peculiar structure. We also found that GC3 in grass lineages is globally evolving very slowly and that the decreasing GC3 gradient observed from 5′ to 3′ along coding sequences is also conserved and ancestral to monocots. This result strongly challenges the previous views on the specificity of grass genomes and we discuss its implications for the possible causes of the evolution of GC content in monocots.

Learning the Kinetics of Amyloid β Pore in Alzheimer's Disease Pathology

Increased synthesis of self-aggregating amyloid beta (Aβ) peptides caused by abnormal processing of amyloid precursor protein (APP) is a hallmark of Alzheimer's disease (AD) pathogenesis. Aβ mediates it's effect by disrupting the integrity of cell membrane and interacting with plasma membrane channels. The soluble form of Aβ aggregates into calcium (Ca2+) - permeable pores in the membrane. Aβ pores promote uncontrolled increase in the cytoplasmic Ca2+ concentration by allowing Ca2+ influx to the cell, in addition to enhancing the activity of Ca2+- permeable channels on the plasma membrane and intracellular compartments. The Ca2+ influx through Aβ pores upsets the otherwise fine-tuned micrometer-sized elementary Ca2+ release events and whole-cell Ca2+ response. The disrupted Ca2+ signaling in turn has the potential to alter cell function in many ways.Here, we have used computational modeling in conjunction with TIRF microscopy to study the function of Aβ pores in AD cells. TIRF microscopy was used to image Ca2+ flux through thousands of Aβ pores in parallel at the millisecond scale and single channel resolution. The fluorescence time-series from individual pores was idealized by extending the Maximum Likelihood-based method developed for separating signal from baseline in noisy quantal data (Bruno et al. 2013, Biophys. J. 105:68). The idealized data was used to developed data-driven models for the kinetics of Aβ pore at different stages of it's life. In addition to providing deep insights into the kinetics of Aβ pore, this study demonstrates that the massive imaging data obtained from thousands of channels in parallel using TIRF microscopy can be utilized for single molecule modeling in the same manner as electrical patch-clamp data. Employing the optical patch-clamp data for Markov chain modeling has the added advantage of the experiments being done under physiological conditions.

Nested Sampling for Bayesian Model Comparison in the Context of Salmonella Disease Dynamics

Understanding the mechanisms underlying the observed dynamics of complex biological systems requires the statistical assessment and comparison of multiple alternative models. Although this has traditionally been done using maximum likelihood-based methods such as Akaike's Information Criterion (AIC), Bayesian methods have gained in popularity because they provide more informative output in the form of posterior probability distributions. However, comparison between multiple models in a Bayesian framework is made difficult by the computational cost of numerical integration over large parameter spaces. A new, efficient method for the computation of posterior probabilities has recently been proposed and applied to complex problems from the physical sciences. Here we demonstrate how nested sampling can be used for inference and model comparison in biological sciences. We present a reanalysis of data from experimental infection of mice with Salmonella enterica showing the distribution of bacteria in liver cells. In addition to confirming the main finding of the original analysis, which relied on AIC, our approach provides: (a) integration across the parameter space, (b) estimation of the posterior parameter distributions (with visualisations of parameter correlations), and (c) estimation of the posterior predictive distributions for goodness-of-fit assessments of the models. The goodness-of-fit results suggest that alternative mechanistic models and a relaxation of the quasi-stationary assumption should be considered.

Resampling Procedures for Making Inference Under Nested Case–Control Studies

The nested case–control (NCC) design has been widely adopted as a cost-effective solution in many large cohort studies for risk assessment with expensive markers, such as the emerging biologic and genetic markers. To analyze data from NCC studies, conditional logistic regression and maximum likelihood-based methods have been proposed. However, most of these methods either cannot be easily extended beyond the Cox model or require additional modeling assumptions. More generally applicable approaches based on inverse probability weighting (IPW) have been proposed as useful alternatives. However, due to the complex correlation structure induced by repeated finite risk set sampling, interval estimation for such IPW estimators remain challenging especially when the estimation involves nonsmooth objective functions or when making simultaneous inferences about functions. Standard resampling procedures such as the bootstrap cannot accommodate the correlation and thus are not directly applicable. In this article, we propose a resampling procedure that can provide valid estimates for the distribution of a broad class of IPW estimators. Simulation results suggest that the proposed procedures perform well in settings when analytical variance estimator is infeasible to derive or gives less optimal performance. The new procedures are illustrated with data from the Framingham Offspring Study to characterize individual level cardiovascular risks over time based on the Framingham risk score, C-reactive protein, and a genetic risk score. Supplementary materials for this article are available online.

A comparison of power analysis methods for evaluating effects of a predictor on slopes in longitudinal designs with missing data

In many longitudinal studies, evaluating the effect of a binary or continuous predictor variable on the rate of change of the outcome, i.e. slope, is often of primary interest. Sample size determination of these studies, however, is complicated by the expectation that missing data will occur due to missed visits, early drop out, and staggered entry. Despite the availability of methods for assessing power in longitudinal studies with missing data, the impact on power of the magnitude and distribution of missing data in the study population remain poorly understood. As a result, simple but erroneous alterations of the sample size formulae for complete/balanced data are commonly applied. These 'naive' approaches include the average sum of squares and average number of subjects methods. The goal of this article is to explore in greater detail the effect of missing data on study power and compare the performance of naive sample size methods to a correct maximum likelihood-based method using both mathematical and simulation-based approaches. Two different longitudinal aging studies are used to illustrate the methods.

Probabilistic correlation of single stratigraphic samples: A generalized approach for biostratigraphic data

Existing quantitative methods for biostratigraphic dating and correlation commonly ignore one of the key strengths of the microfossil record—relative abundance data. In this study, we present a maximum likelihood-based biostratigraphic method that demonstrates how microfossil abundance can be used in the stratigraphic placement of isolated samples. Precise correlation and dating of isolated paleontological samples is not possible with current methods, which are primarily intended for the alignment of longer stratigraphic sequences. In contrast, the probabilistic approach provided by likelihood analysis results in sample age estimates with defined confidence intervals. Therefore, all the uncertainties inherent in our age assessment (resulting from small sample sizes, incomplete sampling, imperfect knowledge of stratigraphic distributions, lack of taxonomic resolution of biostratigraphic data, and underlying environmental, paleogeographic, and sedimentologic processes) are explicit in our results. We conclude with a field test of the method, from data collected from an oil well from the Catatumbo Basin, Colombia, illustrating the use of our approach in a real-world case study and highlighting how our method could be generalized to a wide range of stratigraphic problems.

Determining the evolutionary history of gene families

Recent large-scale studies of individuals within a population have demonstrated that there is widespread variation in copy number in many gene families. In addition, there is increasing evidence that the variation in gene copy number can give rise to substantial phenotypic effects. In some cases, these variations have been shown to be adaptive. These observations show that a full understanding of the evolution of biological function requires an understanding of gene gain and gene loss. Accurate, robust evolutionary models of gain and loss events are, therefore, required. We have developed weighted parsimony and maximum likelihood methods for inferring gain and loss events. To test these methods, we have used Markov models of gain and loss to simulate data with known properties. We examine three models: a simple birth-death model, a single rate model and a birth-death innovation model with parameters estimated from Drosophila genome data. We find that for all simulations maximum likelihood-based methods are very accurate for reconstructing the number of duplication events on the phylogenetic tree, and that maximum likelihood and weighted parsimony have similar accuracy for reconstructing the ancestral state. Our implementations are robust to different model parameters and provide accurate inferences of ancestral states and the number of gain and loss events. For ancestral reconstruction, we recommend weighted parsimony because it has similar accuracy to maximum likelihood, but is much faster. For inferring the number of individual gene loss or gain events, maximum likelihood is noticeably more accurate, albeit at greater computational cost. www.bioinf.manchester.ac.uk/dupliphy simon.lovell@manchester.ac.uk; simon.whelan@manchester.ac.uk Supplementary data are available at Bioinformatics online.

A QTL for genotype by sex interaction for anthropometric measurements in Alaskan Eskimos (GOCADAN Study) on chromosome 19q12-13.

Variation in anthropometric measurements due to sexual dimorphism can be the result of genotype by sex interactions (G×S). The purpose of this study was to examine the sex-specific genetic architecture in anthropometric measurements in Alaskan Eskimos from the Genetics of Coronary Artery Disease in Alaska Natives (GOCADAN) study. Maximum likelihood-based variance components decomposition methods, implemented in SOLAR, were used for G×S analyses. Anthropometric measurements included BMI, waist circumference (WC), waist/height ratio, percent body fat (%BF), and subscapular and triceps skinfolds. Except for WC, mean values of all phenotypes were significantly different in men and women (P < 0.05). All anthropometric measures were significantly heritable (P < 0.001). In a preliminary analysis not allowing for G×S interaction, evidence of linkage was detected between markers D19S414 and D19S220 on chromosome 19 for WC (logarithm of odds (lod) = 3.5), %BF (lod = 1.7), BMI (lod = 2.4), waist/height ratio (lod = 2.5), subscapular (lod = 2.1), and triceps skinfolds (lod = 1.9). In subsequent analyses which allowed for G×S interaction, linkage was again found between these traits and the same two markers on chromosome 19 with significantly improved lod scores for: WC (lod = 4.5), %BF (lod = 3.8), BMI (lod = 3.5), waist/height ratio (lod = 3.2), subscapular (lod = 3.0), and triceps skinfolds (lod = 2.9). These results support the evidence of a G×S interaction in the expression of genetic effects resulting in sexual dimorphism in anthropometric phenotypes and identify the chromosome 19q12-13 region as important for adiposity-related traits in Alaskan Eskimos.

Accurate Estimation of Expression Levels of Homologous Genes in RNA-seq Experiments

Abstract Next generation high-throughput sequencing (NGS) is poised to replace array-based technologies as the experiment of choice for measuring RNA expression levels. Several groups have demonstrated the power of this new approach (RNA-seq), making significant and novel contributions and simultaneously proposing methodologies for the analysis of RNA-seq data. In a typical experiment, millions of short sequences (reads) are sampled from RNA extracts and mapped back to a reference genome. The number of reads mapping to each gene is used as proxy for its corresponding RNA concentration. A significant challenge in analyzing RNA expression of homologous genes is the large fraction of the reads that map to multiple locations in the reference genome. Currently, these reads are either dropped from the analysis, or a naive algorithm is used to estimate their underlying distribution. In this work, we present a rigorous alternative for handling the reads generated in an RNA-seq experiment within a probabilistic model for RNA-seq data; we develop maximum likelihood-based methods for estimating the model parameters. In contrast to previous methods, our model takes into account the fact that the DNA of the sequenced individual is not a perfect copy of the reference sequence. We show with both simulated and real RNA-seq data that our new method improves the accuracy and power of RNA-seq experiments.

Within-Host Dynamics of the Hepatitis C Virus Quasispecies Population in HIV-1/HCV Coinfected Patients

HIV/HCV coinfected individuals under highly active antiretroviral therapy (HAART) represent an interesting model for the investigation of the role played by the immune system in driving the evolution of the HCV quasispecies. We prospectively studied the intra-host evolution of the HCV heterogeneity in 8 coinfected subjects, selected from a cohort of 32 patients initiating HAART: 5 immunological responders (group A) and 3 immunological non-responders (group B), and in two HCV singly infected controls not assuming drugs (group C). For all these subjects at least two serial samples obtained at the first observation (before HAART) and more than 1 year later, underwent clonal sequence analysis of partial E1/E2 sequences, encompassing the whole HVR1. Evolutionary rates, dated phylogenies and population dynamics were co-estimated by using a Bayesian Markov Chain Monte Carlo approach, and site specific selection pressures were estimated by maximum likelihood-based methods. The intra-host evolutionary rates of HCV quasispecies was 10 times higher in subjects treated with HAART than in controls without immunodeficiency (1.9 and 2.3×10−3 sub/site/month in group A and B and 0.29×10−3 sub/site/month in group C individuals). The within-host Bayesian Skyline plot analysis showed an exponential growth of the quasispecies populations in immunological responders, coinciding with a peak in CD4 cell counts. On the contrary, quasispecies population remained constant in group B and in group C controls. A significant positive selection pressure was detected in a half of the patients under HAART and in none of the group C controls. Several sites under significant positive selection were described, mainly included in the HVR1. Our data indicate that different forces, in addition to the selection pressure, drive an exceptionally fast evolution of HCV during HAART immune restoration. We hypothesize that an important role is played by the enlargement of the viral replicative space.