From Empirical Bayes to Full Bayes: Methods for Analysing Traffice Safety Data, October 25, 2004

Traffic safety engineers are among the early adopters of Bayesian statistical tools for analyzing crash data. As in many other areas of application, empirical Bayes methods were their first choice, perhaps because they represent an intuitively appealing, yet relatively easy to implement alternative to purely classical approaches. With the enormous progress in numerical methods made in recent years and with the availability of free, easy to use software that permits implementing a fully Bayesian approach, however, there is now ample justification to progress towards fully Bayesian analyses of crash data. The fully Bayesian approach, in particular as implemented via multi-level hierarchical models, has many advantages over the empirical Bayes approach. In a full Bayesian analysis, prior information and all available data are seamlessly integrated into posterior distributions on which practitioners can base their inferences. All uncertainties are thus accounted for in the analyses and there is no need to pre-process data to obtain Safety Performance Functions and other such prior estimates of the effect of covariates on the outcome of interest. In this slight, fully Bayesian methods may well be less costly to implement and may result in safety estimates with more realistic standard errors. In this manuscript, we present the full Bayesian approach to analyzing traffic safety data and focus on highlighting the differences between the empirical Bayes and the full Bayes approaches. We use an illustrative example to discuss a step-by-step Bayesian analysis of the data and to show some of the types of inferences that are possible within the full Bayesian framework.

From Empirical Bayes to Full Bayes: Methods for Analyzubg Traffic Safey Data, October 25, 2004

Traffic safety engineers are among the early adopters of Bayesian statistical tools for analyzing crash data. As in many other areas of application, empirical Bayes methods were their first choice, perhaps because they represent an intuitively appealing, yet relatively easy to implement alternative to purely classical approaches. With the enormous progress in numerical methods made in recent years and with the availability of free, easy to use software that permits implementing a fully Bayesian approach, however, there is now ample justification to progress towards fully Bayesian analyses of crash data. The fully Bayesian approach, in particular as implemented via multi-level hierarchical models, has many advantages over the empirical Bayes approach. In a full Bayesian analysis, prior information and all available data are seamlessly integrated into posterior distributions on which practitioners can base their inferences. All uncertainties are thus accounted for in the analyses and there is no need to pre-process data to obtain Safety Performance Functions and other such prior estimates of the effect of covariates on the outcome of interest. In this light, fully Bayesian methods may well be less costly to implement and may result in safety estimates with more realistic standard errors. In this manuscript, we present the full Bayesian approach to analyzing traffic safety data and focus on highlighting the differences between the empirical Bayes and the full Bayes approaches. We use an illustrative example to discuss a step-by-step Bayesian analysis of the data and to show some of the types of inferences that are possible within the full Bayesian framework.

DnaSP, DNA polymorphism analyses by the coalescent and other methods

DnaSP is a software package for the analysis of DNA polymorphism data. Present version introduces several new modules and features which, among other options allow: (1) handling big data sets (~5 Mb per sequence); (2) conducting a large number of coalescent-based tests by Monte Carlo computer simulations; (3) extensive analyses of the genetic differentiation and gene flow among populations; (4) analysing the evolutionary pattern of preferred and unpreferred codons; (5) generating graphical outputs for an easy visualization of results. Availability: The software package, including complete documentation and examples, is freely available to academic users from: http://www.ub.es/dnasp

Rapid radiation in spiny lobsters (Palinurus spp) as revealed by classic and ABC methods using mtDNA and microsatellite data

Background: Molecular tools may help to uncover closely related and still diverging species from a wide variety of taxa and provide insight into the mechanisms, pace and geography of marine speciation. There is a certain controversy on the phylogeography and speciation modes of species-groups with an Eastern Atlantic-Western Indian Ocean distribution, with previous studies suggesting that older events (Miocene) and/or more recent (Pleistocene) oceanographic processes could have influenced the phylogeny of marine taxa. The spiny lobster genus Palinurus allows for testing among speciation hypotheses, since it has a particular distribution with two groups of three species each in the Northeastern Atlantic (P. elephas, P. mauritanicus and P. charlestoni) and Southeastern Atlantic and Southwestern Indian Oceans (P. gilchristi, P. delagoae and P. barbarae). In the present study, we obtain a more complete understanding of the phylogenetic relationships among these species through a combined dataset with both nuclear and mitochondrial markers, by testing alternative hypotheses on both the mutation rate and tree topology under the recently developed approximate Bayesian computation (ABC) methods. Results Our analyses support a North-to-South speciation pattern in Palinurus with all the South-African species forming a monophyletic clade nested within the Northern Hemisphere species. Coalescent-based ABC methods allowed us to reject the previously proposed hypothesis of a Middle Miocene speciation event related with the closure of the Tethyan Seaway. Instead, divergence times obtained for Palinurus species using the combined mtDNA-microsatellite dataset and standard mutation rates for mtDNA agree with known glaciation-related processes occurring during the last 2 my. Conclusion The Palinurus speciation pattern is a typical example of a series of rapid speciation events occurring within a group, with very short branches separating different species. Our results support the hypothesis that recent climate change-related oceanographic processes have influenced the phylogeny of marine taxa, with most Palinurus species originating during the last two million years. The present study highlights the value of new coalescent-based statistical methods such as ABC for testing different speciation hypotheses using molecular data.

Data-driven clustering : new methods and applications

Abstract : This work is concerned with the development and application of novel unsupervised learning methods, having in mind two target applications: the analysis of forensic case data and the classification of remote sensing images. First, a method based on a symbolic optimization of the inter-sample distance measure is proposed to improve the flexibility of spectral clustering algorithms, and applied to the problem of forensic case data. This distance is optimized using a loss function related to the preservation of neighborhood structure between the input space and the space of principal components, and solutions are found using genetic programming. Results are compared to a variety of state-of--the-art clustering algorithms. Subsequently, a new large-scale clustering method based on a joint optimization of feature extraction and classification is proposed and applied to various databases, including two hyperspectral remote sensing images. The algorithm makes uses of a functional model (e.g., a neural network) for clustering which is trained by stochastic gradient descent. Results indicate that such a technique can easily scale to huge databases, can avoid the so-called out-of-sample problem, and can compete with or even outperform existing clustering algorithms on both artificial data and real remote sensing images. This is verified on small databases as well as very large problems. Résumé : Ce travail de recherche porte sur le développement et l'application de méthodes d'apprentissage dites non supervisées. Les applications visées par ces méthodes sont l'analyse de données forensiques et la classification d'images hyperspectrales en télédétection. Dans un premier temps, une méthodologie de classification non supervisée fondée sur l'optimisation symbolique d'une mesure de distance inter-échantillons est proposée. Cette mesure est obtenue en optimisant une fonction de coût reliée à la préservation de la structure de voisinage d'un point entre l'espace des variables initiales et l'espace des composantes principales. Cette méthode est appliquée à l'analyse de données forensiques et comparée à un éventail de méthodes déjà existantes. En second lieu, une méthode fondée sur une optimisation conjointe des tâches de sélection de variables et de classification est implémentée dans un réseau de neurones et appliquée à diverses bases de données, dont deux images hyperspectrales. Le réseau de neurones est entraîné à l'aide d'un algorithme de gradient stochastique, ce qui rend cette technique applicable à des images de très haute résolution. Les résultats de l'application de cette dernière montrent que l'utilisation d'une telle technique permet de classifier de très grandes bases de données sans difficulté et donne des résultats avantageusement comparables aux méthodes existantes.

Les méthodes d'études de cas en psychothérapie: perspectives actuelles [Methods of psychotherapy case studies: current perspectives].

There is a renewal of interest among psychotherapy researchers and psychotherapists towards psychotherapy case studies. This article presents two paradigms that have greatly influenced this increasing interest in psychotherapy case studies : the pragmatic case study and the theory-building case study paradigm. The origins, developments and key-concepts of both paradigms are presented, as well as their methodological and ethical specificities. Examples of case studies, along with models developed, are cited. The differential influence of the post-modern schools on both paradigms are presented, as well as their contribution to the field of methods of psychotherapy case studies discussed and assessed in terms of relevance for the researcher and the psychotherapist.

Optimization of analytical methods for the assessment of the quality of fats and oils used in continuous deep fat frying

La aplicabilidad, repetibilidad y capacidad de diferentes métodos de análisis para discriminar muestras de aceites con diferentes grados de oxidación fueron evaluadas mediante aceites recogidos en procesos de fritura en continuo en varias empresas españolas. El objetivo de este trabajo fue encontrar métodos complementarios a la determinación del índice de acidez para el control de calidad rutinario de los aceites de fritura empleados en estas empresas. La optimización de la determinación de la constante dieléctrica conllevó una clara mejora de la variabilidad. No obstante, excepto en el caso del índice del ATB, el resto de métodos ensayados mostraron una menor variabilidad. La determinación del índice del ATB fue descartada ya que su sensibilidad fue insuficiente para discriminar entre aceites con diferente grado de oxidación. Los diferentes parámetros de alteración determinados en los aceites de fritura mostraron correlaciones significativas entre el índice de acidez y varios parámetros de oxidación diferentes, como la constante dieléctrica, el índice de p-anisidina, la absorción al ultravioleta y el contenido en polímeros de los triacilgliceroles. El índice de acidez solo evalúa la alteración hidrolítica, por lo que estos parámetros aportan información complementaria al evaluar la alteración termooxidativa.

The importance of micro-habitat in the breeding of Barn Owls Tyto alba

Capsule Habitat parameters associated with 706 Barn Owl (Tyto alba) nesting boxes in Israel were analysed. Pairs bred in 259 of the boxes. The intensity of agricultural practices at nestbox sites were shown to have only a weak effect on aspects of Barn Owl breeding in this region.

Counterfactuals and causal inference: Methods and principles for social research

"Most quantitative empirical analyses are motivated by the desire to estimate the causal effect of an independent variable on a dependent variable. Although the randomized experiment is the most powerful design for this task, in most social science research done outside of psychology, experimental designs are infeasible. (Winship & Morgan, 1999, p. 659)." This quote from earlier work by Winship and Morgan, which was instrumental in setting the groundwork for their book, captures the essence of our review of Morgan and Winship's book: It is about causality in nonexperimental settings.

Chiroptical measurement of chiral aggregates at liquid-liquid interface in centrifugal liquid membrane cell by Mueller matrix and conventional circular dichroism methods

The centrifugal liquid membrane (CLM) cell has been utilized for chiroptical studies of liquid-liquid interfaces with a conventional circular dichroism (CD) spectropolarimeter. These studies required the characterization of optical properties of the rotating cylindrical CLM glass cell, which was used under the high speed rotation. In the present study, we have measured the circular and linear dichroism (CD and LD) spectra and the circular and linear birefringence (CB and LB) spectra of the CLM cell itself as well as those of porphyrine aggregates formed at the liquid-liquid interface in the CLM cell, applying Mueller matrix measurement method. From the results, it was confirmed that the CLM-CD spectra of the interfacial porphyrin aggregates observed by a conventional CD spectropolarimeter should be correct irrespective of LD and LB signals in the CLM cell.

Characterising avian haemosporidian communities : ecological factors, parasite persistence and co-infection

Les parasites jouent un rôle clef dans l'évolution des comportements et des traits d'histoire de vie de leurs hôtes. Le parasitisme s'avère parfois dévastateur à l'échelle de population d'hôtes, et peut également altérer certains traits associés à la valeur sélective d'un individu infecté, tels que son succès reproducteur ou encore son taux de mortalité. La coévolution hôte/parasite, qui représente l'une des forces sélectives les plus puissantes dans l'évolution des organismes, peut également conduire les partenaires de l'association parasitaire à s'adapter localement à des environnements hétérogènes. Cette thèse porte sur l'étude de parasites aviaires, du genre Plasmodium, Haemopro- teus et Leucocytozoon (Haemosporidae), naturellement associés à différentes populations de mésanges charbonnières (Parus major) et d'hirondelles des fenêtres (Delichon ur- bicum). Dans un premier temps, nous avons cherché à déterminer comment se distribuent ces parasites au sein de différentes populations hôtes et si ces communautés de parasites sont structurées. Par la suite, la principale question à laquelle nous voulions répondre était de savoir comment ces parasites, et notamment après coexistence de plusieurs lignées génétiques d'Haemosporidae au sein dun même-individu (i.e. co-infection), affectent la physiologie et le succès de reproducteur des hôtes. Nos résultats suggèrent que la distribution des Haemosporidae est principalement gouvernée par la présence d'insectes vecteurs et que la persistance de l'infection chez les hôtes varie en fonction du genre d'Haemosporidae (Chapitre 1-2). Par ailleurs, nous avons trouvé que des lignées de parasite génétiquement distinctes peuvent avoir des effets contrastés sur leurs hôtes. Par exemple, les hôtes exhibent des différences de parasitémie marquées en fonction des lignées de parasites responsable de l'infection. De plus, le succès reproducteur ainsi que la charge parasitaire des mésanges infectées par Plasmodium ou Haemoproteus n'étaient pas affecté par l'infection simultanée avec Leucocytozoon (Chapitre 2-3). Dans le Chapitre 4, j'ai examiné la capacité immunitaire de mésanges charbonnières infectées par des hémosporidies. Les résultats n'ont pas été concluant, et je suggère fortement une réévaluation de ceux-ci dans de futures études. Les mésanges charbonnières ne semblent pas signaler leur statut infectieux par la coloration de leur plumage (Chapitre 5); toutefois, la coloration noire des plumes reflète l'état de stress oxydatif des mésanges, qui dépend lui-même de l'infection parasitaire. La coloration verte pourrait également indiquer la qualité des soins paxentaux délivrés par les mésanges adultes femelles à leurs petits, comme le suggère la corrélation que nous avons observée entre la masse des jeunes d'une nichée et la coloration de leur mère. Les hirondelles capturées en Algérie souffrent plus de l'infection que celles échantillon¬nées en Europe (Chapitre 6). Les similitudes observées entre les communautés de par¬asites affectant les populations européennes et celles des populations nord-africaines suggèrent que la transmission des parasites a lieu lors de la migration vers le sud. A l'instar de nos observations sur les mésanges dans les chapitres 2 et 3, les hirondelles co-infectées ne montrent pas d'altérations de leur condition physique. Cette thèse démontre qu'il existe, au sein des populations de mésanges charbonnières, des interactions antagonistes entre, d'une part, les parasites et leurs hôtes et d'autre part, entre différent parasites. Le résultat de ces interactions antagonistes varie en fonction des espèces et de la zone géographique considérée. Nous avons démontré que les interactions ne suivent pas toujours la théorie, puisque la coevolution qui, en suivant le concept de la virulence, devrait augmenter la charge parasitaire et diminuer la condition physique des hôtes, ne montre pourtant pas d'impact négatif sur les populations de mésanges. Nous pouvons maintenant concentrer nos efforts à la caractérisation des interactions antagonistes. De plus, grâce aux avancées des méthodes moléculaires, nous pouvons suivre et étudier en détails comment ces interactions se manifestent et quels sont leurs effets sur la condition physique des hôtes. - Parasites are key in shaping various behavioural and life-history traits of their hosts. The influence of parasitism on host populations varies from slight to devastating and might influence such parameters as mortality rates or reproductive success. Host-parasite coevolution is one of the most powerful selective forces in evolution and can lead to local adaptation of parasites and hosts in spatially structured environments. In this thesis, I studied haemosporidian parasites in different populations of great tits (Parus major) and house martins (Delichon urbicum). Firstly, I wanted to determine how parasites are distributed and if parasite communities are structured. The main question I wanted to address hereafter was how parasites, and specifically infection with multiple genera of parasites (i.e. co-infection) influenced host physiology and reproductive success. I found that parasite distribution is environmentally driven and could therefore be closely linked to vector prevalence; and that the stability of parasite infection over time is genus-dependent (Chapter 1 - 2). I further found that different haemosporidian lineages might interact differently with their hosts as parasitaemia was strongly lineage-specific and that the presence of Leucocytozoon parasites showed no correlation to Plasmodium or Haemoproteus parasitaemia, nor to great tit reproductive success (Chapter 2-3). In Chapter 4 I examined immune capacity of haemosporidian-infected great tits. The results proved inconclusive, and I strongly suggest re-evaluation hereof in future work. Great tits do not appear to signal parasite infection through plumage colouration (Chapter 5); however, infection did have a link to oxidative stress resistance which is strongly signalled through the black breast stripe, with darker males being more resistant and darker females less resistant. Females might incur different costs associated with darker stripes. This would allow reversal of signaling function. Green colouration could also serve as a cue for female provisioning quality as indicated by the strong correlation between colouration and chick body mass. Breeding house martins caught in Algeria suffer greater haemosporidian infection than European populations (Chapter 6). Similar parasite communities in European and North-African populations suggest transmission of parasites may occur during southward migration. Similarly to what was observed in great tits in Chapter 2 and 3, no relationship was found between parasite co-infection and Swiss house martin body condition. This thesis demonstrates that host-parasite and inter-parasite antagonistic interac¬tions exist in great tit populations. How these interactions play out is species dependent and varies geographically. I have demonstrated that interactions do not always follow the theory, as co-infection - which under the concept of virulence should increase parasitaemia and decrease body condition - showed no negative impact on great tit populations. We can now concentrate our efforts on characterising these antagonistic interactions, and with the advance in molecular methods, track and investigate how these interactions play out and what the effect on host fitness is.

Differences in diet between the two largest breeding colonies of Audouin's gulls: the effects of fishery activities

We analysed and compared the diet of Audouin´s gulls Larus audouinii between their two largest breeding sites in the world: the Ebro Delta and the Chafarinas Islands (western Mediterranean). These two localities showed marked differences in the features of the commercial fishing fleet: in the Ebro Delta area a large fishing fleet produced large amounts of discards, while in the Chafarinas the fleet discarded smaller amounts of fish and marine invertebrates, due to the smaller number of vessels. It is also likely that the percentage of discards from total catches is also lower around the Chafarinas than at the Ebro Delta. We distinguished two types of fishing to compare diet compositions: diurnal (only trawling activity) and diurnal and nocturnal (trawling and purse-seine activity, respectively). We also differentiated regurgitates from young nestlings (up to 20 days old) and from older nestlings or adult birds. At the two localities, fish was the main food of Audouin´s gulls, with epipelagic prey (mainly clupeoids) being more important when both diurnal and nocturnal fisheries were operating. This confirms that epipelagic prey either caught actively by the gulls or linked to fisheries was particularly important in the feeding habits of Audouin´s gulls. Nevertheless, differences between the two colonies appear mainly when only trawlers operated: while at the Ebro Delta gulls showed higher consumption of benthic-mesopelagic prey (probably linked to a higher trawler discard availability), gulls from the Chafarinas Islands consumed higher biomass of epipelagic prey probably caught actively at night. When both fleets operated around the two colonies, the average biomass of prey in a regurgitate of younger chicks was significantly higher at the Ebro Delta than at Chafarinas, and the opposite trend was recorded for older nestlings and adults. Niche width was broader in Chafarinas than in the Ebro Delta for both age classes and for any fishing fleet schedule, suggesting again that the exploitation of discards was higher at the Ebro Delta than at the Chafarinas, where gulls showed a more varied diet. Despite the fact that availability of discards was probably higher at the Ebro Delta than at Chafarinas, the per capita availability was not so different at both localities due to the increasing seabird community population at the Ebro Delta, which ca. doubled that at Chafarinas in the last decade.