Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “statistical inference”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,369 records · Page 76Linked to original sources
PubMed2004

Multilevel linear modelling for FMRI group analysis using Bayesian inference.

Functional magnetic resonance imaging studies often involve the acquisition of data from multiple sessions and/or multiple subjects. A hierarchical approach can be taken to modelling such data with a general linear model (GLM) at each level of the hierarchy introducing different random effects variance components. Inferring on these models is nontrivial with frequentist solutions being unavailable. A solution is to use a Bayesian framework. One important ingredient in this is the choice of prior on the variance components and top-level regression parameters. Due to the typically small numbers of sessions or subjects in neuroimaging, the choice of prior is critical. To alleviate this problem, we introduce to neuroimage modelling the approach of reference priors, which drives the choice of prior such that it is noninformative in an information-theoretic sense. We propose two inference techniques at the top level for multilevel hierarchies (a fast approach and a slower more accurate approach). We also demonstrate that we can infer on the top level of multilevel hierarchies by inferring on the levels of the hierarchy separately and passing summary statistics of a noncentral multivariate t distribution between them.

Bayes Theorem↗
PubMed2004

Mechanisms of theory formation in young children.

Research suggests that by the age of five, children have extensive causal knowledge, in the form of intuitive theories. The crucial question for developmental cognitive science is how young children are able to learn causal structure from evidence. Recently, researchers in computer science and statistics have developed representations (causal Bayes nets) and learning algorithms to infer causal structure from evidence. Here we explore evidence suggesting that infants and children have the prerequisites for making causal inferences consistent with causal Bayes net learning algorithms. Specifically, we look at infants and children's ability to learn from evidence in the form of conditional probabilities, interventions and combinations of the two.

Bayes Theorem↗
PubMed2005

Use of AFLP markers in surveys of arthropod diversity.

Arthropods comprise the most diverse group of animals on earth and as such have been the subject of considerable evolutionary research. For example, much of our understanding of the genetic basis of evolutionary change is derived from the insect genus Drosophila, one of the most well-studied organisms in biology. Arthropods are also of tremendous economic importance as both providers and chief destroyers of food for human consumption. Thus, the genetic diversity of arthropods is of interest from both a pure research perspective and for practical economic reasons. The amplified fragment length polymorphism (AFLP) method of genetic analysis, developed in the early and mid-1990s (Vos et al., 1995; Zabeau, 1992; Zabeau and Vos, 1993), offers a relatively new method for assessing genetic diversity and has been increasingly applied in studies of arthropods. Originally coined selective restriction fragment amplification (SRFA) (Zabeau and Vos, 1993), the method was renamed (Vos et al., 1995) presumably to reflect its similarity to restriction fragment length polymorphism (RFLP). Since then, AFLPs have become a popular tool in both population genetics to estimate population parameters such as heterozygosity, F-statistics, migration rates, and genetic distances, as well as phylogenetics, to infer relationships among closely related taxa. In arthropods, AFLPs have been used to assess genetic variation both within and between species in various taxa including crustaceans, chelicerates, and insects, often yielding novel insights. In this chapter, we briefly describe the AFLP method and its strengths and limitations. We then discuss the use of AFLPs in surveys of arthropod diversity, highlighting the specific questions addressed using AFLPs. Finally, a section on experimental design and methods, based on research in our laboratory, is provided.

Animals↗
PubMed1998

Mentalism, mechanisms, and medical analogues: reply to Wakefield (1998).

J. C. Wakefield's (1998) critique of W. C. Follette and A. C. Houts's (1996) article is addressed by raising questions about (a) mentalism as a framework for studying psychopathology, (b) the nature of inferred mechanisms and the process of making such inferences, and (c) the accuracy of claiming that the Diagnostic and Statistical Manual of Mental Disorders (DSM; American Psychiatric Association) expansion is analogous to the success of physical medicine. The authors argue that Wakefield's analysis of mental disorders leads to the conclusion that the modern DSMs are not reflective of the progress of physical medicine.

Humans↗
PubMed1999

Assessment of human exposure to ambient particulate matter.

Recent epidemiological studies have consistently shown that the acute mortality effects of high concentrations of ambient particulate matter (PM), documented in historic air pollution episodes, may also be occurring at the low to moderate concentrations of ambient PM found in modern urban areas. In London in December 1952, the unexpected deaths due to PM exposure could be identified and counted as integers by the coroners. In modern times, the PM-related deaths cannot be as readily identified, and they can only be inferred as fractional average daily increases in mortality rates using sophisticated statistical filtering and analyses of the air quality and mortality data. The causality of the relationship between exposure to ambient PM and acute mortality at these lower modern PM concentrations has been questioned because of a perception that there is little significant correlation in time between the ambient PM concentrations and measured personal exposure to PM from all sources (ambient PM plus indoor-generated PM). This article shows that the critical factor supporting the plausibility of a linear PM mortality relationship is the expected high correlation in time of people's exposure to PM of ambient origin with measured ambient PM concentrations, as used in the epidemiological time series studies. The presence of indoor and personal sources of PM masks this underlying relationship, leading to confusion in the scientific literature about the strong underlying temporal relationship between personal exposure to PM of ambient origin and ambient PM concentration. The authors show that the sources of PM of non-ambient origin operate independently of the ambient PM concentrations, so that the mortality effect of non-ambient PM, if any, must be independent of the effects of the ambient PM exposures.

Air Pollutants, Occupational↗
PubMed2003

Greedily building protein networks with confidence.

MOTIVATION: With genome sequences complete for human and model organisms, it is essential to understand how individual genes and proteins are organized into biological networks. Much of the organization is revealed by proteomics experiments that now generate torrents of data. Extracting relevant complexes and pathways from high-throughput proteomics data sets has posed a challenge, however, and new methods to identify and extract networks are essential. We focus on the problem of building pathways starting from known proteins of interest. RESULTS: We have developed an efficient, greedy algorithm, SEEDY, that extracts biologically relevant biological networks from protein-protein interaction data, building out from selected seed proteins. The algorithm relies on our previous study establishing statistical confidence levels for interactions generated by two-hybrid screens and inferred from mass spectrometric identification of protein complexes. We demonstrate the ability to extract known yeast complexes from high-throughput protein interaction data with a tunable parameter that governs the trade-off between sensitivity and selectivity. DNA damage repair pathways are presented as a detailed example. We highlight the ability to join heterogeneous data sets, in this case protein-protein interactions and genetic interactions, and the appearance of cross-talk between pathways caused by re-use of shared components. SIGNIFICANCE AND COMPARISON: The significance of the SEEDY algorithm is that it is fast, running time O[(E + V) log V] for V proteins and E interactions, a single adjustable parameter controls the size of the pathways that are generated, and an associated P-value indicates the statistical confidence that the pathways are enriched for proteins with a coherent function. Previous approaches have focused on extracting sub-networks by identifying motifs enriched in known biological networks. SEEDY provides the complementary ability to perform a directed search based on proteins of interest. AVAILABILITY: SEEDY software (Perl source), data tables and confidence score models (R source) are freely available from the author.

Algorithms↗
PubMed1999

Incongruence in the diploid B-genome species complex of Glycine (Leguminosae) revisited: histone H3-D alleles versus chloroplast haplotypes.

Variation at the single-copy nuclear locus histone H3-D was surveyed in the diploid B-genome group of Glycine subgenus Glycine (Leguminosae: Papilionoideae), which comprises three named Australian species and a number of distinct but as yet not formally recognized taxa. A total of 23 alleles was identified in the 44 accessions surveyed. Only one individual was clearly heterozygous, which is not surprising given the largely autogamous breeding system of subgenus Glycine. Alleles differed by as many as 19 nucleotide substitutions, nearly all in the three introns; length variation was minimal. Phylogenetic analysis identified two shortest allele trees with very little homoplasy, suggesting that recombination has been rare. Both topological and data set incongruence were statistically significant between histone H3-D allele trees and trees inferred from chloroplast DNA haplotypes previously described from these same accessions. Whereas the distribution of H3-D alleles agrees well with morphologically based taxonomic groupings, chloroplast DNA haplotype polymorphisms transgress species boundaries, suggesting that the chloroplast genome is not tracking taxic relationships. Divergences among chloroplast DNA haplotypes involved in such transgressive patterns appear to be more recent than speciation events, suggesting hybridization rather than lineage sorting.

Alleles↗
PubMed1997

Comparison of the accuracy of two tests with a confirmatory procedure limited to positive results.

Sensitivity (S) and specificity (F) of a test can be estimated in a study only if the disease status of all subjects is independently confirmed using a gold standard. A confirmatory procedure, however, rarely is administered to all subjects, and studies comparing two tests often restrict confirmation to subjects classified as positive by either test. In this context, the parameters of interest (S1, S2, F1, and F2) cannot be estimated directly. Two useful comparative measures of accuracy that can be estimated in this design are the relative sensitivity (RSN, S1/S2) and the relative false-positive rate [RFP, (1 - F1)/(1 - F2)]. The comparison of the two tests can be regarded as a self-matched study in which each observation consists of a pair of test results. Common statistical techniques for matched data analysis can be used to make inferences about RSN and REP. A variance estimator developed for the log-risk ratio in a matched cohort study can be applied to estimate confidence intervals for RSN and RFP. In addition, conditional logistic regression is useful to test hypotheses about RSN and RFP and to assess the impact of multiple potential modifiers of RSN and RFP.

Adult↗
PubMed1982

Cutaneous mechanoreceptors in macaque monkey: temporal discharge patterns evoked by vibration, and a receptor model.

1. Vibratory stimuli applied to the hand of a monkey evoke phase locked impulse trains in the three classes of low threshold mechanoreceptive afferents which innervate the area. The responses of each class of afferent (slowly adapting (SA), rapidly adapting (RA), and Pacinian (PC) vary in a systematic but complex way across the range of frequencies and intensities to which they are sensitive. The receptors are not accessible for electrophysiological recording. The aim in this study was to infer the mechanisms underlying their responses from detailed examination of the statistical properties of the impulse trains.2. A very simple receptor model with four degrees of freedom was chosen as a starting point. The independent variables consisted of the resting membrane time constant, tau, a variable membrane conductance, G(r), the fraction of each sinusoidal stimulus cycle producing depolarization, p(r), and the noise level, sigma, which was assigned to the impulse threshold. The aim was to use the deviations between observed data and predictions from the basic model to construct a more effective model. In fact, the deviations were minor and were mostly explained by periods of increased excitability in the wake of each action potential. Almost all of the differences between the responses of the three mechanoreceptive classes examined in this paper were accounted for by differences in time constants.3. The temporal structure of the responses from each mechanoreceptive class was examined at two levels of resolution, a coarse level where the resolution unit was a full cycle, and a fine level where the unit was 0.1 ms.4. The coarse structure of each response was represented by the presence or absence of an impulse on each stimulus cycle. In each mechanoreceptive class, the impulse sequences were random at low stimulus frequencies and regular at high frequencies. The transition frequencies were roughly 5 Hz for the slowly adapting afferents, 7 Hz for the rapidly adapting afferents, and 110 Hz for the Pacinian afferents. The model matched these data closely when the time constants were set at 80, 60 and 3.4 ms for SAs, RAs and PCs, respectively.5. The fine structure of the responses of each mechanoreceptive class exhibited impulse phase locking, phase advance with increasing intensity, and bimodal phase distributions at higher frequencies. Impulses contributing to the first mode of bimodal distributions always occurred in cycles following cycles in which no impulse occurred. Impulses contributing to the retarded mode always occurred in cycles following filled cycles. The mean phase differences between the two modes was called phase retardation. Phase retardation grew with stimulus frequency for both the receptors and the model; the time constants required to match the model against neural phase retardation curves were 123, 64 and 4.8 ms for SAs, RAs and PCs, respectively.

Animals↗
PubMed1979

Analysis of circadian periodicity of plasma cortisol in normal man and in Cushing's syndrome.

In normal subjects, adjusted to a circadian environment, half-hourly plasma cortisol levels over a 24-h span (time series, TS) outline a circadian sinusoidal pattern that may be masked by superimposed short-lived oscillations. The periodogram objectively defines the features (amplitude, phase) of this pattern in individual TS and, more efficiently, in the average and concatenation of homogeneous TS. Concatenation is a procedure by which homogeneous short TS are attached one after the other. Homogeneity is inferred by variance analysis and by a new measure, h, with statistical significance, derived from the spectrum of the concatenated TS. Through concatenation, autocovariance for periodicity identification becomes possible. The TS from three groups of volunteers are highly homogeneous and have a predominant, statistically significant amplitude with peak time near 1100 h. The patterns from patients with Cushing's disease differ markedly in their means and periodic features; by the above criteria, we identify a subgroup with significant h measure and 24-h amplitude. This analysis provides a tool for investigation of the circadian clock in man.

Adolescent↗
PubMed2001

Preprocessing implementation for microarray (PRIM): an efficient method for processing cDNA microarray data.

cDNA microarray technology is useful for systematically analyzing the expression profiles of thousands of genes at once. Although many useful results inferred by using this technology and a hierarchical clustering method for statistical analysis have been confirmed using other methods, there are still questions about the reproducibility of the data. We have therefore developed a data processing method that very efficiently extracts reproducible data from the result of duplicate experiments. It is designed to automatically filter the raw results obtained from cDNA microarray image-analysis software. We optimize the threshold value for filtering the data by using the product of N and R, where N is the ratio of the number of spots that passed the filtering vs. the total number of spots, and R is the correlation coefficient for results obtained in the duplicate experiments. Using this method to process mouse tissue expression profile data that contain 1,881,600 points of analysis, we obtained clustered results more reasonable than those obtained using previously reported filtering methods.

Animals↗
PubMed2002

Toward a comprehensive developmental model for major depression in women.

OBJECTIVE: Major depression is a multifactorial disorder with many etiologic variables that are interrelated through developmental pathways. The authors used structural equation modeling to generate a developmental model for the etiology of major depression in women. METHOD: Data from 1,942 adult female twins, interviewed up to four times over a 9-year period, were used to construct a developmental model to predict depressive episodes in the year before the most recent interview. Eighteen risk factors in five developmental tiers were considered: 1) childhood (genetic risk, disturbed family environment, childhood sexual abuse, and childhood parental loss), 2) early adolescence (neuroticism, self-esteem, and early-onset anxiety and conduct disorder), 3) late adolescence (educational attainment, lifetime traumas, social support, and substance misuse), 4) adulthood (history of divorce and past history of major depression), and 5) the last year (marital problems, difficulties, and stressful life events). RESULTS: The best fitting model included six correlations and 64 paths, provided an excellent fit to the data, and explained 52% of the variance in liability to episodes of major depression. The findings suggest that the development of risk for major depression in women results from three broad pathways reflecting internalizing symptoms, externalizing symptoms, and psychosocial adversity. CONCLUSIONS: Major depression is an etiologically complex disorder, the full understanding of which will require consideration of a broad array of risk factors from multiple domains. These results, while plausible, should be treated with caution because of problems with causal inference, retrospective recall bias, and the limitations of a purely additive statistical model.

Adult↗
PubMed2006

Dementia--epidemiological considerations, nomenclature, and a tacit consensus definition.

Epidemiologic inquiry requires the definition of a "case." Dementia may be defined clinically or alternatively by inference of irreversible brain pathology. Several iterations of the Diagnostic and Statistical Manual of Mental Disorders and International Classification of Diseases have skirted this issue by using criteria that are at once syndromic and neuropathological. The limitations of this compromise are revealed by large discrepancies in case identification when the various published criteria are strictly applied. Despite this problem, neuroepidemiologists have produced convergent estimates of the prevalence and incidence of dementia and its association with risk factors. This progress has reflected the tacit reliance of investigators on a simple definition of dementia as the syndrome of substantial global cognitive decline not attributable to alteration in level of consciousness. Beyond this description, our knowledge of pathology and, ultimately, the etiology of individual cases is extremely variable. Whatever its antecedents, syndromically defined dementia presents a looming public health crisis.

Aged↗
PubMed2006

Promoter prediction and annotation of microbial genomes based on DNA sequence and structural responses to superhelical stress.

BACKGROUND: In our previous studies, we found that the sites in prokaryotic genomes which are most susceptible to duplex destabilization under the negative superhelical stresses that occur in vivo are statistically highly significantly associated with intergenic regions that are known or inferred to contain promoters. In this report we investigate how this structural property, either alone or together with other structural and sequence attributes, may be used to search prokaryotic genomes for promoters. RESULTS: We show that the propensity for stress-induced DNA duplex destabilization (SIDD) is closely associated with specific promoter regions. The extent of destabilization in promoter-containing regions is found to be bimodally distributed. When compared with DNA curvature, deformability, thermostability or sequence motif scores within the -10 region, SIDD is found to be the most informative DNA property regarding promoter locations in the E. coli K12 genome. SIDD properties alone perform better at detecting promoter regions than other programs trained on this genome. Because this approach has a very low false positive rate, it can be used to predict with high confidence the subset of promoters that are strongly destabilized. When SIDD properties are combined with -10 motif scores in a linear classification function, they predict promoter regions with better than 80% accuracy. When these methods were tested with promoter and non-promoter sequences from Bacillus subtilis, they achieved similar or higher accuracies. We also present a strictly SIDD-based predictor for annotating promoter sequences in complete microbial genomes. CONCLUSION: In this report we show that the propensity to undergo stress-induced duplex destabilization (SIDD) is a distinctive structural attribute of many prokaryotic promoter sequences. We have developed methods to identify promoter sequences in prokaryotic genomes that use SIDD either as a sole predictor or in combination with other DNA structural and sequence properties. Although these methods cannot predict all the promoter-containing regions in a genome, they do find large sets of potential regions that have high probabilities of being true positives. This approach could be especially valuable for annotating those genomes about which there is limited experimental data.

Bacillus subtilis↗
PubMed1985

Discrimination of luminance increments and decrements.

Discrimination of pulsed increments from decrements improves relative to detection when the adaptation level is decreased or stimulus duration is increased. Using Tanner's theory of recognition [J. Acoust. Soc. Am. 28, 882 (1956)], we show that the subject's internal decision variables corresponding to the increment and decrement stimuli also show increased statistical independence as adaptation is decreased or stimulus duration increases. We infer from this result that at high adaptation levels detection is based predominantly on information in one channel, whereas at low adaptation levels there is information from two channels. Similarly, for long durations, we infer two independent channels, which contribute little information about the sign of the stimulus at short durations.

Discrimination, Psychological↗
PubMed2000

[Clinical trials: principles of the method].

Comparative judgement, which is seminal to any kind of science performing measurements, has been applied to clinical reasoning for many centuries. The need for systematizing the observational methods used in medicine in order to draw more reliable inferences about the effects of therapies has been active all along the 19th century. This has resulted in controlled studies which yielded important advances in clinical and therapeutic knowledge, although their designs were not fully satisfactory. Clinical trials have gained their status of "hard science", methodology allowing causal inference, by the end of the 1940s after having adopted the statistical theories developed in the 1930s by Fisher for experimental design in agronomy. A long way has been run since the first controlled randomized trial. However, half a century later, modern clinical trial remains essentially a controlled randomized prospective study using methods to limit potential biases and to establish statistical significance.

Clinical Trials as Topic↗
PubMed2005

[Effects of maternal infection with Trypanosoma cruzi in pregnancy development and in the newborn infant].

In the endemic regions of Bolivia the infection of the feminine population in fertile age by T. cruzi is frequent (20 to 50 % of the women in fertile age) and the rate of fetal maternal transmission is of approximately 5%. A great percentage of infected women do not transmit the infection to the fetus. The intention of the present study carried out at the Maternal-Infantile Hospital Germán Urquidi of Cochabamba (Bolivia) is to contribute to the knowledge regarding the pregnancy and birth of a newborn of Chagas infected women who do not transmit the infection to the fetus. 2124 mothers and 2,155 newborns were studied. The prevalence of infection by T. cruzi among these pregnant women is of 26,3%. Two groups of mothers were studied: 554 that presented infection by T. cruzi (group M+B-) and 1520 not infected (group control M-B-). Both groups of mothers are comparable in their anthropometric and obstetrical antecedents. The mothers (M+B+) are in average older than those not infected (p<0.05), which will probably have an influence on the number of gestations and abortion antecedents, which were of p<0.05 and p=0.01 respectively. Among the different anthropometric and biological parameters studied in newborns of groups M+B- and M-B -, no statistically significant differences between both groups were found. It can be inferred that the chronic maternal infection by T. cruzi seems to have no clinical influence, neither on the course of the pregnancy nor during birth, if a group of T. cruzi infected mothers is compared to a non infected group.

Adult↗
PubMed2005

General anesthesia: as a challenge and treatment need option in pediatric dentistry.

OBJECTIVE: The purpose of this study was to determine the type and characteristics of the interventions, indications of dental treatment and procedures performed to patients treated under general anesthesia (GA) by pediatric dentistry residents, during the 1997-1999 period. METHOD: A sample of 57 hospital records of patients treated as part of the Special Pediatric Course at the Puerto Rico Pediatric Hospital were reviewed. Statistical analysis was done using the chi-square test for inferences on proportions. RESULTS: MR patients made up 59.7% and NMR patients made up 40.3% of the sample studied. Ages ranged from 2 to 35 years with a mean age of 11 years (SD=8.54). MR patients were classified into 7 categories: mental retardation (38.2%), cerebral palsy (14.7%), epileptic (5.9%), mental syndromes (26.9%), hydrocephalic (5.9%), autism (5.9%) and others (2.9%). The NMR were classified into 5 categories: early childhood caries (65.2%), cardiac patients (8.7%), maxillofacial anomalies (4.3%), organic syndromes (13.1%) and others (8.7%). The dental procedures performed were: dental extractions 84%(MR) and 68% (NMR), restorative procedures 87.3%(MR) and 12.7%(NMR). Oral prophylaxis was performed in 76.8%, fissure sealants in 10.7% and topical fluoride applications in 21.8%. CONCLUSION: Dental extraction was a frequently performed procedure in both groups. The prevalence of exodontia and restorative procedures indicates the need to design and implement prevention programs for special pediatric patients.

Adolescent↗