Biomedical subjects

Stefan Van Dongen

Publications and source records attributed to Stefan Van Dongen.

7 recordsLinked to original sources

PubMed2006

Prior specification in Bayesian statistics: three cautionary tales.

One of the most important differences between Bayesian and traditional techniques is that the former combines information available beforehand-captured in the prior distribution and reflecting the subjective state of belief before an experiment is carried out-and what the data teach us, as expressed in the likelihood function. Bayesian inference is based on the combination of prior and current information which is reflected in the posterior distribution. The fast growing implementation of Bayesian analysis techniques can be attributed to the development of fast computers and the availability of easy to use software. It has long been established that the specification of prior distributions should receive a lot of attention. Unfortunately, flat distributions are often (inappropriately) used in an automatic fashion in a wide range of types of models. We reiterate that the specification of the prior distribution should be done with great care and support this through three examples. Even in the absence of strong prior information, prior specification should be done at the appropriate scale of biological interest. This often requires incorporation of (weak) prior information based on common biological sense. Very weak and uninformative priors at one scale of the model may result in relatively strong priors at other levels affecting the posterior distribution. We present three different examples intuïvely illustrating this phenomenon indicating that this bias can be substantial (especially in small samples) and is widely present. We argue that complete ignorance or absence of prior information may not exist. Because the central theme of the Bayesian paradigm is to combine prior information with current data, authors should be encouraged to publish their raw data such that every scientist is able to perform an analysis incorporating his/her own (subjective) prior distributions.

Animals↗

PubMed2005

High variation in developmental instability under non-normal developmental error: a Bayesian perspective.

The developmental mechanisms behind developmental instability (DI) are only poorly understood. Nevertheless, fluctuating asymmetry (FA) is often used a surrogate for DI. Based on statistical arguments it is often assumed that individual levels of FA are only weakly associated with the underlying DI. Patterns in FA therefore need to be interpreted with caution, and should ideally be transformed into patterns in DI. In order to be able to achieve that, assumptions about the distribution of developmental errors must be made. Current models assume that errors during development are additive and independent such that they yield a normal distribution. The observation that the distribution of FA is often leptokurtic has been interpreted as evidence for between-individual variation in DI. This approach has led to unrealistically high estimates of between-individual variation in DI, and potentially incorrect interpretations of patterns in FA, especially at the individual level. Recently, it has been suggested that the high estimates of variation in DI may be biased upward because either developmental errors are log-normal or gamma distributed and/or low measurement resolution of FA. A proper estimation of the amount (and shape) of heterogeneity in DI is crucial for the interpretation of patterns in FA and their transformation into patterns in DI. Yet, incorrect model assumptions may render misleading inferences. We therefore develop a statistical model to evaluate the sensitivity of results under the normal error model against the two alternative distributions as well as to investigate the importance of low measurement resolution. An analysis of simulated and empirical data sets indicated that bias due to misspecification of the developmental error distribution can be substantial, yet, did not appear to reduce estimates of variation in DI in empirical data sets to a large extent. Effects of low measurement resolution were neglectable. The importance of these results are discussed in the context of the interpretation of patterns in FA.

Animals↗

PubMed2005

Risperidone in children with disruptive behavior disorders and subaverage intelligence: a 1-year, open-label study of 504 patients.

OBJECTIVE: To determine the long-term safety and effectiveness of risperidone for severe disruptive behaviors in children. METHOD: A multisite, 1-year, open-label study of patients aged 5 to 14 years with disruptive behaviors and subaverage intelligence was conducted. RESULTS: Seventy-three percent of the 504 patients enrolled completed the study. The mean +/- SE dose of risperidone was 1.6 +/- 0.0 mg/day. The most common adverse events were somnolence (30%), rhinitis (27%), and headache (22%). The incidence of movement disorders was low, and mean Extrapyramidal Symptom Rating Scale scores decreased during risperidone treatment. No clinically significant changes in mean laboratory values were noted, except for a transient increase in serum prolactin levels. Scores on the Nisonger Child Behavior Rating Form Conduct Problem Scale improved significantly as early as week 1, and improvement was maintained throughout the trial (p < .001 at each time point). Significant improvements were noted on positive social behavior and other Nisonger Child Behavior Rating Form subscales, Aberrant Behavior Checklist, Clinical Global Impressions scale, and tests of patients' cognitive function (each p < .001). CONCLUSIONS: Risperidone was well tolerated and effective in the long-term treatment of disruptive behavior disorders in children with subaverage intelligence.

Adolescent↗

PubMed2003

Variable responses to large-scale climate change in European Parus populations.

Spring temperatures in temperate regions have increased over the past 20 years and many organisms have responded to this increase by advancing the timing of their growth and reproduction. However, not all populations show an advancement of phenology. Understanding why some populations advance and others do not will give us insight into the possible constraints and selection pressures on the advancement of phenology. By combining two decades of data on 24 populations of tits (Parus sp.) from six European countries, we show that the phenological response to large-scale changes in spring temperature varies across a species' range, even between populations situated close to each other. We show that this variation cannot be fully explained by variation in the temperature change during the pre- and post-laying periods, as recently suggested. Instead, we find evidence for a link between rising temperatures and the frequency of second broods, which results in complex shifts in the laying dates of first clutches. Our results emphasize the need to consider links between different life-history parameters in order to predict the ecological consequences of large-scale climate changes.

Adaptation, Physiological↗

PubMed2002

Avian persistence in fragmented rainforest.

What factors determine the persistence of species in fragmented habitats? To address this question, we studied the relative impacts of forest deterioration and fragmentation on bird species in 12 rainforest fragments in Kenya, combining 6 years of individual capture-recapture data with measurements of current captures and museum specimens. Species mobility, as estimated from species-specific dispersal rates, and tolerance to habitat deterioration, as estimated from change in fluctuating asymmetry with increasing habitat disturbance, explained 88% of the variation in patch occupancy among eight forest bird species. Occupancy increased with mobility and with tolerance to deterioration, where both variables contributed equally to this relationship. We conclude that individual-level study, such as of dispersal behavior and phenotypic development, can predict patterns of persistence at the species level. More generally, for conservation tactics to stand a high chance of success, they should include action both within sites, to minimize habitat deterioration, and across landscapes, to maximize dispersal.

Animals↗

PubMed2002

Modelling developmental instability as the joint action of noise and stability: a Bayesian approach.

BACKGROUND: Fluctuating asymmetry is assumed to measure individual and population level developmental stability. The latter may in turn show an association with stress, which can be observed through asymmetry-stress correlations. However, the recent literature does not support an ubiquitous relationship. Very little is known why some studies show relatively strong associations while others completely fail to find such a correlation. We propose a new Bayesian statistical framework to examine these associations RESULTS: We are considering developmental stability - i.e. the individual buffering capacity - as the biologically relevant trait and show that (i) little variation in developmental stability can explain observed variation in fluctuating asymmetry when the distribution of developmental stability is highly skewed, and (ii) that a previously developed tool (i.e. the hypothetical repeatability of fluctuating asymmetry) contains only limited information about variation in developmental stability, which stands in sharp contrast to the earlier established close association between the repeatability and developmental instability. CONCLUSION: We provide tools to generate valuable information about the distribution of between-individual variation in developmental stability. A simple linear transformation of a previous model lead to completely different conclusions. Thus, theoretical modelling of asymmetry and stability appears to be very sensitive to the scale of inference. More research is urgently needed to get better insights in the developmental mechanisms of noise and stability. In spite of the fact that the model is likely to represent an oversimplification of reality, the accumulation of new insights could be incorporated in the Bayesian statistical approach to obtain more reliable estimation.

Bayes Theorem↗

PubMed2002

Fluctuating asymmetry as an indicator of fitness: can we bridge the gap between studies?

There is growing evidence from both experimental and non-experimental studies that fluctuating asymmetry does not consistently index stress or fitness. The widely held--yet poorly substantiated--belief that fluctuating asymmetry can act as a universal measure of developmental stability and predictor of stress-mediated changes in fitness, therefore staggers. Yet attempts to understand why the reported relationships between fluctuating asymmetry, stress and fitness are so heterogeneous--i.e. whether the associations are truly weak or non-existent or whether they become confounded during different stages of the analytical pathways remain surprisingly scarce. Hence, we attempt to disentangle these causes, by reviewing the various statistical and conceptual factors that are suspected to confound potential relationships between fluctuating asymmetry, stress and fitness. Two main categories of factors are discerned: those associated with the estimation of developmental stability through fluctuating asymmetry and those associated with the effects of genotype and environment on developmental stability. Next, we describe a series of statistical tools that have recently been developed to help reduce this noise. We argue that the current lack of a theoretical framework that predicts if and when relationships with developmental stability can be expected, urges for further theoretical and empirical research, such as on the genetic architecture of developmental stability in stressed populations. If the underlying developmental mechanisms are better understood, statistical patterns of asymmetry variation may become a biologically meaningful tool.

Animals↗