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.

SourceResearch connectionFrom yearThrough year

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

At least 577 records · Page 32Linked to original sources

PubMed2002

A computational perspective on the neural basis of multisensory spatial representations.

We argue that current theories of multisensory representations are inconsistent with the existence of a large proportion of multimodal neurons with gain fields and partially shifting receptive fields. Moreover, these theories do not fully resolve the recoding and statistical issues involved in multisensory integration. An alternative theory, which we have recently developed and review here, has important implications for the idea of 'frame of reference' in neural spatial representations. This theory is based on a neural architecture that combines basis functions and attractor dynamics. Basis function units are used to solve the recoding problem, whereas attractor dynamics are used for optimal statistical inferences. This architecture accounts for gain fields and partially shifting receptive fields, which emerge naturally as a result of the network connectivity and dynamics.

Animals↗

PubMed2002

Use of the Self-Administered Eczema Area and Severity Index by parent caregivers: results of a validation study.

BACKGROUND: The Eczema Area and Severity Index (EASI) is used by dermatological investigators world-wide to assess eczema disease severity. EASI measures are, however, time-consuming and require trained personnel, thereby limiting its application to large-scale epidemiological studies. Additionally, the use of self-assessed severity indices in dermatology is restricted to adult subjects and conditions, thereby not addressing the needs of paediatric patients. OBJECTIVES: To develop and validate an instrument for a caregiver's self-assessment of the severity of his/her child's atopic dermatitis (AD), the Self-Administered EASI (SA-EASI). METHODS: Trained investigators performed a modified EASI assessment on the same day as an SA-EASI was obtained from 47 caregivers of children with AD. RESULTS: The SA-EASI was found to be a valid measure of the severity of AD. Total, acute and chronic SA-EASI scores predicted total, acute and chronic modified EASI scores (P < 0.0001). SA-EASI body surface area (BSA) scores predicted EASI BSA scores (P < 0.0001). SA-EASI pruritus scores correlated with the acute, chronic and total EASI scores (P = 0.0001). CONCLUSIONS: The SA-EASI may provide caregivers the means to report the severity of their child's skin disease objectively. The high correlation with the EASI score observed in this sample implies that statistical inferences with the SA-EASI will be valid for large populations. In future studies, this will permit analysis of the relationship of skin disease severity to such measures as quality of life, disability, patient satisfaction and the costs of various therapies. Moreover, this SA-EASI instrument may allow older children, over 12 years old, to assess the severity of their AD.

Acute Disease↗

PubMed2001

Effects of thiazide- and loop-diuretics, alone or in combination, on calcitropic hormones and biochemical bone markers: a randomized controlled study.

OBJECTIVE: Diuretics are commonly used drugs that in addition to their effect on the cardiovascular system also affect calcium homeostasis and bone metabolism. We evaluated the effects of loop diuretics (LD) and thiazide diuretics (TD) on calcitropic hormones and biochemical bone markers. DESIGN: A total of 50 postmenopausal women were randomized to 7 days of treatment with either the TD bendroflumethiazide, the LD bumetanide, bendroflumethiazide plus bumetanide, or placebo. Blood and urine (24 h) were sampled on each day. Statistical inferences were made versus the concomitant changes in the placebo group. RESULTS: Bendroflumethiazide increased the tubular reabsorption of calcium (TRCa) (+0.46 +/- 0.11%, P=0.009), plasma levels of parathyroid hormones (PTH) (+24 +/- 10%, P=0.06), and 1,25(OH)2D (+12 +/- 6%, P=0.03). Bumetanide decreased the TRCa (-0.5 +/- 0.1%, P=0.01) and increased plasma PTH and 1,25(OH)2D levels (+27 +/- 9%, P=0.02 and +36 +/- 12%, P=0.006, respectively). Treatment with either of the drugs did not alter plasma calcium, osteocalcin, bone alkaline phosphatase (bone-ALP) or urinary NTx/creatinine ratio. However, treatment with both drugs caused an increased plasma calcium level (+2.7 +/- 1.0%, P=0.007) and decreased plasma levels of bone-ALP (-21 +/- 3%, P=0.001), osteocalcin (-6 +/- 3%, P=0.03), and urinary NTx/creatinine ratio (-39 +/- 6%, P=0.001). CONCLUSION: Calcium homeostasis and bone metabolism are to a major degree influenced by diuretic treatment. Surprisingly, LD and TD exerted a similar effect on calcitropic hormones despite their opposite effects on the renal calcium excretion. In clinical practice, treatment with diuretics has to be considered as a cause of parathyroid stimulation.

Aged↗

PubMed1997

Nonlinear deterministic analysis of tissue texture: a stereological study on mastopathic and mammary cancer tissue using chaos theory.

Signals from some dynamical systems look like stochastic processes although their latent mechanism is deterministic. When such systems show sensitive dependence on small changes of initial conditions, they are denoted as deterministic chaos. Motivated by recent advances in statistical inference methods for chaotic systems and by the concept of spatial chaos, we present a deterministic approach to the study of epithelial tissue texture. Methods for estimation of the autocorrelation function, for evaluation of the power spectrum, for attractor reconstruction, for estimation of the Lyapunov exponent and of the correlation dimension, and for the generation of surrogate data sets are outlined. In our biological example, these methods are applied to 20 cases of mastopathy as compared to 20 cases of mammary cancer. The input signals for the analysis were estimates of epithelial fraction measured at low magnification within 5100 equally spaced line segments per case perpendicular to an arbitrarily directed axis. The results suggest the existence of a low-dimensional deterministic attractor in mastopathic tissue texture, which is replaced by coloured noise in the majority of mammary carcinomas. Biological mechanisms for this finding and scale effects are discussed, and some methodological aspects and possible extensions of our approach are outlined.

Breast Neoplasms↗

PubMed1999

Evidence for a major gene accounting for mild elevation in LDL cholesterol: the NHLBI Family Heart Study.

Studies of rare Mendelian disorders of low density lipoprotein cholesterol (LDL-C) metabolism have identified specific genetic mutations in the LDL receptor and apolipoprotein B. Although these rare mutations account for a small proportion of LDL-C variation, twin and adoption studies indicate that at least 50% of the overall LDL-C observed variation is genetically determined. In a heterogeneous sample of 3227 subjects from the NHLBI Family Heart Study collected from four US centres, we find evidence for a common major gene accounting for mild elevations (1.25 standard deviations) in LDL-C. The analysis favored a recessive model with a frequency of 0.52 for the gene influencing elevated LDL-C, phenotypic means of 113 mg/dl for the normal genotypes and 146 mg/dl for the abnormal genotype, and a significant polygenic heritability. This statistically-inferred major gene accounted for 24% of the variation in LDL-C, with polygenes accounting for another 28% of the variation. Using parameters for major gene transmission estimated in the segregation analysis, LDL-C showed no linkage to the LDL receptor gene (LDLR), nor to the apolipoprotein E gene (APOE), nor to the cholesterol 7alpha-hydroxylase gene (CYP7A1), indicating the major gene effect influencing mild elevation in LDL-C is not explained by any of these candidate loci.

Adult↗

PubMed2004

Tests of gene-environment interaction for case-parent triads with general environmental exposures.

As knowledge of the human genome continues to grow, more progress is being made towards not only identifying the genes involved in disease susceptibility but also in defining the synergistic role genes play with environmental exposures. The detection of gene-environment interactions is important as it can offer clinicians a potential means of intervention. The discovery of interactions relies heavily on powerful statistical methods. We present a test, FBAT-I, that can be used to investigate gene-environment interaction. The test uses the case-parent triad design and protects the statistical inference from potential spurious results due to population admixture.

Environmental Exposure↗

PubMed2006

Single and multiple dose bioequivalence evaluation of two brands of gliclazide modified release tablets in healthy Chinese male volunteers.

Randomized, two-way, crossover, single- and multiple-dose studies were conducted in healthy Chinese male volunteers to evaluate the bioequivalence of two brands of gliclazide (CAS 21187-98-4, 1-(3-azabicylco(3, 3, 0)oct-3-yl)-3-p-tolysulfonylurea) 30 mg tablets, viz. Gliclazide modified release (MR) tablets as test (T) and a commercial gliclazide standard preparation as reference (R) product. Each volunteer received T and R tablets separated by 7 days of a drug-free washout period. The plasma concentrations of gliclazide, determined by a validated LC-ESI-MS method, were employed to assess the pharmacokinetic parameters such as maximum and minimum observed plasma concentration (Cmax and Cmin), time to Cmax (tmax), average plasma concentration at steady state (Cav), area under plasma concentration curve (AUC(0-72), AUC(0-infinity) and AUC(ss), and degree of fluctuation for plasma concentration (DF %). As to these parameters, the analysis of variance (ANOVA) showed no significant difference and 90 % confidence intervals (CI) fell entirely into the acceptable range of bioequivalence. Based on these statistical inferences, the two formulations are considered bioequivalent in the extent and rate of absorption from both single- and multiple-dose studies.

Adult↗

PubMed2000

Mixed model analysis of quantitative trait loci.

We develop a mixed model approach of quantitative trait locus (QTL) mapping for a hybrid population derived from the crosses of two or more distinguished outbred populations. Under the mixed model, we treat the mean allelic value of each source population as the fixed effect and the allelic deviations from the mean as random effects so that we can partition the total genetic variance into between- and within-population variances. Statistical inference of the QTL parameters is obtained by using the Bayesian method implemented by Markov chain Monte Carlo (MCMC). This unified QTL mapping algorithm treats the fixed and random model approaches as special cases of the general mixed model methodology. Utility and flexibility of the method are demonstrated by using a set of simulated data.

Algorithms↗

PubMed2000

The asymptotic distribution of canonical correlations and variates in cointegrated models.

The cointegrated model considered here is a nonstationary vector autoregressive process in which some linear functions are stationary and others are random walks. The first difference of the process (the "error-correction form") is stationary. Statistical inference, such as reduced rank regression estimation of the coefficients of the process and tests of hypotheses of dimensionality of the stationary part, involves the canonical correlations between the difference vector and the relevant vector of the past of the process. The asymptotic distributions of the canonical correlations and the canonical vectors under the assumption that the process is Gaussian are found.

Journal Article↗

PubMed2000

Effect of neighboring bases on base-pair stacking orientation: a molecular dynamics study.

It is generally believed that base-pair stacking interaction in DNA double helix is one of the strongest interactions that governs sequence directed structural variability. However, X-ray crystal structures of some base-paired doublet sequences have been seen to adopt different structures when flanked by different base-pairs. DNA crystal database, however, is still too small to make good statistical inference about effect of such flanking residues. Influence of neighboring residue on the local helical geometry of a base-paired doublet in B-DNA has been investigated here using molecular dynamics simulation. We have generated ensembles of structures for d(CA).d(TG) and d(AA).d(TT) base-paired doublets located at the centers of d(CGCGCAAAGCG).d(CGCTTTGCGCG) and d(CGCGAAAACGCG).d(CGCGTTTTCGCG) sequences along with their analogs by varying the bases either at 5'- or 3'- position to the central doublet. Comparison of base paired doublet parameters for the ensembles of structures show that stacking geometry of d(CA).d(TG) doublet depends on some of the flanking base-pairs. On the other hand d(AA).d(TT) doublet remains nearly unperturbed when the flanking residues are altered.

Base Pairing↗

PubMed2005

Genetic analysis and attribution of microbial forensics evidence.

Because of the availability of pathogenic microorganisms and the relatively low cost of preparing and disseminating bioweapons, there is a continuing threat of biocrime and bioterrorism. Thus, enhanced capabilities are needed that enable the full and robust forensic exploitation and interpretation of microbial evidence from acts of bioterrorism or biocrimes. To respond to the need, greater resources and efforts are being applied to the burgeoning field of microbial forensics. Microbial forensics focuses on the characterization, analysis and interpretation of evidence for attributional purposes from a bioterrorism act, biocrime, hoax or inadvertent agent release. To enhance attribution capabilities, a major component of microbial forensics is the analysis of nucleic acids to associate or eliminate putative samples. The degree that attribution can be addressed depends on the context of the case, the available knowledge of the genetics, phylogeny, and ecology of the target microorganism, and technologies applied. The types of genetic markers and features that can impact statistical inferences of microbial forensic evidence include: single nucleotide polymorphisms, repetitive sequences, insertions and deletions, mobile elements, pathogenicity islands, virulence and resistance genes, house keeping genes, structural genes, whole genome sequences, asexual and sexual reproduction, horizontal gene transfer, conjugation, transduction, lysogeny, gene conversion, recombination, gene duplication, rearrangements, and mutational hotspots. Nucleic acid based typing technologies include: PCR, real-time PCR, MLST, MLVA, whole genome sequencing, and microarrays.

Bacterial Typing Techniques↗

PubMed1981

A conversational information computer system for health and safety operation: the occupational surveillance interactive system (OSIS).

Collection, storage and retrieval of health and safety information is critical in occupational settings in making decisions about safety of employees and the healthful nature of the work environment. Computer surveillance programs have been utilized previously, but have been too complicated, too mathematical, and not directly accessible to health and safety personnel. This paper discusses the Occupational Surveillance Interactive System (OSIS), a conversational system which utilizes English instead of a computer language, and which provides immediate feedback to the health and safety practitioner through fact retrieval and statistical inference. Additionally, an algorithm is introduced which provides a predictive cueing for the development of some disease processes. The hypothetical example illustrated herein introduces a prototypal computer surveillance system, operating under an interactive mode, to provide the health and safety professional with a decision-making tool in occupational settings.

Computers↗

PubMed2002

Groundhog Day: cause and effect and the primary importance of the finite population induced by randomization.

The logical structure that enables randomized clinical trials to establish cause and effect is reviewed. Statisticians have a major role to define this structure clearly and to help clinicians apply statistical inference in ways directly related to their trials. Scientific importance of establishing limited cause and effect should be accented rather than inferences to generalized populations beyond the scope of a trial's actual random sampling. Authors of clinical reports should clearly define the randomization-induced populations to which their inferences apply. Careful attention to identifying this population of inference can lead to resolution of some issues commonly debated in the analysis of clinical trials. If formal inferences to generalized populations are attempted, these populations should also be carefully described, the assumed random sampling process should be carefully defined, and appropriate methods should be used that correspond with the assumed random sampling process.

Humans↗

PubMed2005

Anomalous anticipatory skin conductance response to acoustic stimuli: experimental results and speculation about a mechanism.

OBJECTIVES: The primary aim of this study was to conduct a replication, simplification, and extension of similar previous studies that claimed anomalous anticipatory skin conductance responses prior to various stimuli and to provide sufficient protocol and analysis details in order to foster additional replications. A secondary aim was to provide a testable model in order to understand the observed results. DESIGN: We used standard skin conductance measures and techniques to search within 50 participants for responses prior to 1-second duration, 97-dB acoustic stimuli, compared to prior to silent controls. We used an interstimulus interval randomly and uniformly distributed between 30 and 50 seconds. OUTCOME MEASURES: The dependent variable was the difference between proportions of 3.5-second prestimulus intervals prior to acoustic stimuli and prior to silent controls that contained a fully formed, nonspecific skin conductance response (ns-SCR). The null hypothesis was that the proportion difference should be zero. RESULTS: We found a significant proportion difference of 0.032 (Z = 2.08; effect size = 0.077 +/- 0.037; p(1t) = 0.0018), which is a replication of earlier similar studies. CONCLUSIONS: We examined and ruled out a number of potential artifacts that might have accounted for this finding. To understand these results, we demonstrated, by Monte Carlo techniques, that a possible explanation is that experimenters may have used their own intuition to initiate experiment runs to somehow sort otherwise random nonspecific skin conductance responses into appropriate bins in order to mimic physiological responses. We found experimental evidence to support this idea as an operational mechanism. If this speculation is confirmed in prospective studies, then this intuition-based mimicking of effects may profoundly impact the interpretation of results from complementary and alternative medical studies that use statistical inference to assess outcomes.

Acoustic Stimulation↗

PubMed2005

Pseudo-likelihood analysis of codon substitution models with neighbor-dependent rates.

Recently, Markov processes for the evolution of coding DNA with neighbor dependence in the instantaneous substitution rates have been considered. The neighbor dependency makes the models analytically intractable, and previously Markov chain Monte Carlo methods have been used for statistical inference. Using a pseudo-likelihood idea, we introduce in this paper an approximative estimation method which is fast to compute. The pseudo-likelihood estimates are shown to be very accurate, and from analyzing 348 human-mouse coding sequences we conclude that the incorporation of a CpG effect improves the fit of the model considerably.

Algorithms↗

PubMed2006

Modular DAG-RNN architectures for assembling coarse protein structures.

We develop and test machine learning methods for the prediction of coarse 3D protein structures, where a protein is represented by a set of rigid rods associated with its secondary structure elements (alpha-helices and beta-strands). First, we employ cascades of recursive neural networks derived from graphical models to predict the relative placements of segments. These are represented as discretized distance and angle maps, and the discretization levels are statistically inferred from a large and curated dataset. Coarse 3D folds of proteins are then assembled starting from topological information predicted in the first stage. Reconstruction is carried out by minimizing a cost function taking the form of a purely geometrical potential. We show that the proposed architecture outperforms simpler alternatives and can accurately predict binary and multiclass coarse maps. The reconstruction procedure proves to be fast and often leads to topologically correct coarse structures that could be exploited as a starting point for various protein modeling strategies. The fully integrated rod-shaped protein builder (predictor of contact maps + reconstruction algorithm) can be accessed at http://distill.ucd.ie/.

Algorithms↗

PubMed2005

Research findings and strategies for assessing telemedicine costs.

With varying degrees of enthusiasm, researchers and decision-makers support the use of telemedicine. Forms of telemedicine are appearing in health-care delivery, and are often integral to transforming health-care information technology. Despite this, the appropriate role of telemedicine in the delivery process remains ambiguous, at least partly because of its uncertain impact on costs. Cost savings and benefits are often suggested by the logic of its impact on health care and by the promise of technology, but definitive information on the costs and benefits remain elusive. The objectives of this paper are to review the state of telemedicine cost research, to examine major issues affecting the yield from this research, and finally to recommend strategies for improving future research. As this paper demonstrates, the productivity of telemedicine cost studies suffers from an under-utilization of appropriate program evaluation and economic methods. This review of telemedicine cost literature will appraise telemedicine cost studies and their findings within a broad analytic framework. Telemedicine cost studies will be assessed on their methods of statistical inference, use of critical economic concepts, and contextual definition for the determination of costs and benefits.

Cost-Benefit Analysis↗

PubMed2003

Whole-proteome interaction mining.

MOTIVATION: A major post-genomic scientific and technological pursuit is to describe the functions performed by the proteins encoded by the genome. One strategy is to first identify the protein-protein interactions in a proteome, then determine pathways and overall structure relating these interactions, and finally to statistically infer functional roles of individual proteins. Although huge amounts of genomic data are at hand, current experimental protein interaction assays must overcome technical problems to scale-up for high-throughput analysis. In the meantime, bioinformatics approaches may help bridge the information gap required for inference of protein function. In this paper, a previously described data mining approach to prediction of protein-protein interactions (Bock and Gough, 2001, Bioinformatics, 17, 455-460) is extended to interaction mining on a proteome-wide scale. An algorithm (the phylogenetic bootstrap) is introduced, which suggests traversal of a phenogram, interleaving rounds of computation and experiment, to develop a knowledge base of protein interactions in genetically-similar organisms. RESULTS: The interaction mining approach was demonstrated by building a learning system based on 1,039 experimentally validated protein-protein interactions in the human gastric bacterium Helicobacter pylori. An estimate of the generalization performance of the classifier was derived from 10-fold cross-validation, which indicated expected upper bounds on precision of 80% and sensitivity of 69% when applied to related organisms. One such organism is the enteric pathogen Campylobacter jejuni, in which comprehensive machine learning prediction of all possible pairwise protein-protein interactions was performed. The resulting network of interactions shares an average protein connectivity characteristic in common with previous investigations reported in the literature, offering strong evidence supporting the biological feasibility of the hypothesized map. For inferences about complete proteomes in which the number of pairwise non-interactions is expected to be much larger than the number of actual interactions, we anticipate that the sensitivity will remain the same but precision may decrease. We present specific biological examples of two subnetworks of protein-protein interactions in C. jejuni resulting from the application of this approach, including elements of a two-component signal transduction systems for thermoregulation, and a ferritin uptake network.

Algorithms↗