Aspectos computacionais para inferência na distribuição gama generalizada

Pós-graduação em Matematica Aplicada e Computacional - FCT

Comparison of a genetic group and unknown paternity models for growth traits in Nellore cattle

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Genetic association between milk yield, stayability, and mastitis in Holstein cows under tropical conditions

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Genetic parameter estimates for feed efficiency and dry matter intake and their association with growth and carcass traits in Nellore cattle

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Genetic parameters and trends of morphometric traits of GIFT tilapia under selection for weight gain

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Construção de uma escala para avaliação da pro ciência em conteúdos matemáticos básicos

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Modelos lineares generalizados bayesianos para dados longitudinais

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

A multiple time scale survival model with a cure fraction

Many recent survival studies propose modeling data with a cure fraction, i.e., data in which part of the population is not susceptible to the event of interest. This event may occur more than once for the same individual (recurrent event). We then have a scenario of recurrent event data in the presence of a cure fraction, which may appear in various areas such as oncology, finance, industries, among others. This paper proposes a multiple time scale survival model to analyze recurrent events using a cure fraction. The objective is analyzing the efficiency of certain interventions so that the studied event will not happen again in terms of covariates and censoring. All estimates were obtained using a sampling-based approach, which allows information to be input beforehand with lower computational effort. Simulations were done based on a clinical scenario in order to observe some frequentist properties of the estimation procedure in the presence of small and moderate sample sizes. An application of a well-known set of real mammary tumor data is provided.

Genetic diversity and population structure of Musa accessions in ex situ conservation

Abstract Background Banana cultivars are mostly derived from hybridization between wild diploid subspecies of Musa acuminata (A genome) and M. balbisiana (B genome), and they exhibit various levels of ploidy and genomic constitution. The Embrapa ex situ Musa collection contains over 220 accessions, of which only a few have been genetically characterized. Knowledge regarding the genetic relationships and diversity between modern cultivars and wild relatives would assist in conservation and breeding strategies. Our objectives were to determine the genomic constitution based on Internal Transcribed Spacer (ITS) regions polymorphism and the ploidy of all accessions by flow cytometry and to investigate the population structure of the collection using Simple Sequence Repeat (SSR) loci as co-dominant markers based on Structure software, not previously performed in Musa. Results From the 221 accessions analyzed by flow cytometry, the correct ploidy was confirmed or established for 212 (95.9%), whereas digestion of the ITS region confirmed the genomic constitution of 209 (94.6%). Neighbor-joining clustering analysis derived from SSR binary data allowed the detection of two major groups, essentially distinguished by the presence or absence of the B genome, while subgroups were formed according to the genomic composition and commercial classification. The co-dominant nature of SSR was explored to analyze the structure of the population based on a Bayesian approach, detecting 21 subpopulations. Most of the subpopulations were in agreement with the clustering analysis. Conclusions The data generated by flow cytometry, ITS and SSR supported the hypothesis about the occurrence of homeologue recombination between A and B genomes, leading to discrepancies in the number of sets or portions from each parental genome. These phenomenons have been largely disregarded in the evolution of banana, as the “single-step domestication” hypothesis had long predominated. These findings will have an impact in future breeding approaches. Structure analysis enabled the efficient detection of ancestry of recently developed tetraploid hybrids by breeding programs, and for some triploids. However, for the main commercial subgroups, Structure appeared to be less efficient to detect the ancestry in diploid groups, possibly due to sampling restrictions. The possibility of inferring the membership among accessions to correct the effects of genetic structure opens possibilities for its use in marker-assisted selection by association mapping.

A novel map-matching procedure for low-sampling GPS data with applications to traffic flow analysis

An extensive sample (2%) of private vehicles in Italy are equipped with a GPS device that periodically measures their position and dynamical state for insurance purposes. Having access to this type of data allows to develop theoretical and practical applications of great interest: the real-time reconstruction of traffic state in a certain region, the development of accurate models of vehicle dynamics, the study of the cognitive dynamics of drivers. In order for these applications to be possible, we first need to develop the ability to reconstruct the paths taken by vehicles on the road network from the raw GPS data. In fact, these data are affected by positioning errors and they are often very distanced from each other (~2 Km). For these reasons, the task of path identification is not straightforward. This thesis describes the approach we followed to reliably identify vehicle paths from this kind of low-sampling data. The problem of matching data with roads is solved with a bayesian approach of maximum likelihood. While the identification of the path taken between two consecutive GPS measures is performed with a specifically developed optimal routing algorithm, based on A* algorithm. The procedure was applied on an off-line urban data sample and proved to be robust and accurate. Future developments will extend the procedure to real-time execution and nation-wide coverage.

Phylogeny of Western Palaearctic long-eared bats (Mammalia, Chiroptera, Plecotus) - a molecular perspective

Die Phylogenie der Westpaläarktischen Langohren (Mammalia, Chiroptera, Plecotus) – eine molekulare Analyse Die Langohren stellen eine Fledermausgattung dar, die fast alle westpaläarktischen Habitate bist zum Polarkreis hin besiedeln und in vielerlei Hinsicht rätselhaft sind. In der Vergangenheit wurden zahlreiche Formen und Varietäten beschrieben. Trotzdem galt für lange Zeit, dass nur zwei Arten in Europa anerkannt wurden. Weitere Arten waren aus Nordafrika, den Kanaren und Asien bekannt, aber auch deren Artstatus wurde vielfach in Frage gestellt. In der vorliegenden Dissertation habe ich mittels molekularer Daten,der partiellen Sequenzierung mitochondrialer Gene (16S rRNA und ND1), sowie der mitochondrialen Kontrollregion, eine molekular Analyse der phylogenetischen Verwandtschaftsverhältnisse innerhalb und zwischen den Linien der westpaläarktischen Langohren durchgeführt. Die besten Substitutionsmodelle wurden berechnet und phylogenetische Bäume mit Hilfe vier verschiedener Methoden konstruiert: dem neighbor joining Verfahren (NJ), dem maximum likelihood Verfahren (ML), dem maximum parsimony Verfahren (MP) und dem Bayesian Verfahren. Sechs Linien der Langohren sind genetisch auf einem Artniveau differenziert: Plecotus auritus, P. austriacus, P. balensis, P. christii, P. sardus, und P. macrobullaris. Im Falle der Arten P. teneriffae, P. kolombatovici und P. begognae ist die alleinige Interpretation der genetischen Daten einzelner mitochondrialer Gene für eine Festlegung des taxonomischen Ranges nicht ausreichend. Ich beschreibe in dieser Dissertation drei neue Taxa: Plecotus sardus, P. kolombatovici gaisleri (=Plecotus teneriffae gaisleri, Benda et al. 2004) and P. macrobullaris alpinus [=Plecotus alpinus, Kiefer & Veith 2002). Morphologische Kennzeichen, insbesondere für die Erkennung im Feld, werden hier dargestellt. Drei der sieben Arten sind polytypisch: P. auritus (eine west- und ein osteuropäische Linie, eine sardische Linie und eine aktuell entdeckte kaukasische Linie, Plecotus kolombatovici (P. k. kolombatovici, P. k. gaisleri und P. k. ssp.) und P. macrobullaris (P. m. macrobullaris und P. m. alpinus). Die Verbreitungsgebiete der meisten Arten werden in dieser Arbeit erstmals ausschließlich anhand genetisch zugeordneter Tiere dargestellt.Die Untersuchung der ökologischen Einnischung der nun anerkannten Formen, insbesondere in Gebieten sympatrischer Verbreitung, bietet ein spannendes und lohnendes Feld für zukünftige Forschungen. Nicht zuletzt muss sich die Entdeckung eines beachtlichen Anteils kryptischer Diversität innerhalb der westpaläarktischen Langohren auch bei der Entwicklung spezieller Artenschutzkonzepte widerspiegeln.

Sviluppo di un approccio congiunto fuzzy-Bayesiano per l'analisi e la modellizzazione degli ecosistemi: applicazione ad ecosistemi marini costieri.

Nell’attuale contesto di aumento degli impatti antropici e di “Global Climate Change” emerge la necessità di comprenderne i possibili effetti di questi sugli ecosistemi inquadrati come fruitori di servizi e funzioni imprescindibili sui quali si basano intere tessiture economiche e sociali. Lo studio previsionale degli ecosistemi si scontra con l’elevata complessità di questi ultimi in luogo di una altrettanto elevata scarsità di osservazioni integrate. L’approccio modellistico appare il più adatto all’analisi delle dinamiche complesse degli ecosistemi ed alla contestualizzazione complessa di risultati sperimentali ed osservazioni empiriche. L’approccio riduzionista-deterministico solitamente utilizzato nell’implementazione di modelli non si è però sin qui dimostrato in grado di raggiungere i livelli di complessità più elevati all’interno della struttura eco sistemica. La componente che meglio descrive la complessità ecosistemica è quella biotica in virtù dell’elevata dipendenza dalle altre componenti e dalle loro interazioni. In questo lavoro di tesi viene proposto un approccio modellistico stocastico basato sull’utilizzo di un compilatore naive Bayes operante in ambiente fuzzy. L’utilizzo congiunto di logica fuzzy e approccio naive Bayes è utile al processa mento del livello di complessità e conseguentemente incertezza insito negli ecosistemi. I modelli generativi ottenuti, chiamati Fuzzy Bayesian Ecological Model(FBEM) appaiono in grado di modellizare gli stati eco sistemici in funzione dell’ elevato numero di interazioni che entrano in gioco nella determinazione degli stati degli ecosistemi. Modelli FBEM sono stati utilizzati per comprendere il rischio ambientale per habitat intertidale di spiagge sabbiose in caso di eventi di flooding costiero previsti nell’arco di tempo 2010-2100. L’applicazione è stata effettuata all’interno del progetto EU “Theseus” per il quale i modelli FBEM sono stati utilizzati anche per una simulazione a lungo termine e per il calcolo dei tipping point specifici dell’habitat secondo eventi di flooding di diversa intensità.

The evolution of massive clumps in star forming regions

In this thesis two related arguments are investigated: - The first stages of the process of massive star formation, investigating the physical conditions and -properties of massive clumps in different evolutionary stages, and their CO depletion; - The influence that high-mass stars have on the nearby material and on the activity of star formation. I characterise the gas and dust temperature, mass and density of a sample of massive clumps, and analyse the variation of these properties from quiescent clumps, without any sign of active star formation, to clumps likely hosting a zero-age main sequence star. I briefly discuss CO depletion and recent observations of several molecular species, tracers of Hot Cores and/or shocked gas, of a subsample of these clumps. The issue of CO depletion is addressed in more detail in a larger sample consisting of the brightest sources in the ATLASGAL survey: using a radiative tranfer code I investigate how the depletion changes from dark clouds to more evolved objects, and compare its evolution to what happens in the low-mass regime. Finally, I derive the physical properties of the molecular gas in the photon-dominated region adjacent to the HII region G353.2+0.9 in the vicinity of Pismis 24, a young, massive cluster, containing some of the most massive and hottest stars known in our Galaxy. I derive the IMF of the cluster and study the star formation activity in its surroundings. Much of the data analysis is done with a Bayesian approach. Therefore, a separate chapter is dedicated to the concepts of Bayesian statistics.

Statistical mapping of real environments with uwb radars

In this master thesis I evaluated the performance of a Ultra-Wide Bandwidth (UWB) radar system for indoor environments mapping. In particular, I used a statistical Bayesian approach which is able to combine all the measurements collected by the radar, including system non-idealities such as the error on the estimated antenna pointing direction or on the estimated radar position. First I verified through simulations that the system was able to provide a sufficiently accurate reconstruction of the surrounding environment despite the limitations imposed by the UWB technology. In fact, the emission of UWB pulses is limited in terms of transmitted power by international regulations. Motivated by the promising results obtained through simulations, I successively carried out a measurement campaign in a real indoor environment using a UWB commercial device. The obtained results showed that the UWB radar system is capable of providing an accurate reconstruction of indoor environments also adopting not directional antennas.

Genetic diversity among horse populations with a special focus on the Franches-Montagnes breed

Genetic characterization helps to assure breed integrity and to assign individuals to defined populations. The objective of this study was to characterize genetic diversity in six horse breeds and to analyse the population structure of the Franches-Montagnes breed, especially with regard to the degree of introgression with Warmblood. A total of 402 alleles from 50 microsatellite loci were used. The average number of alleles per locus was significantly lower in Thoroughbreds and Arabians. Average heterozygosities between breeds ranged from 0.61 to 0.72. The overall average of the coefficient of gene differentiation because of breed differences was 0.100, with a range of 0.036-0.263. No significant correlation was found between this parameter and the number of alleles per locus. An increase in the number of homozygous loci with increasing inbreeding could not be shown for the Franches-Montagnes horses. The proportion of shared alleles, combined with the neighbour-joining method, defined clusters for Icelandic Horse, Comtois, Arabians and Franches-Montagnes. A more disparate clustering could be seen for European Warmbloods and Thoroughbreds, presumably from frequent grading-up of Warmbloods with Thoroughbreds. Grading-up effects were also observed when Bayesian and Monte Carlo resampling approaches were used for individual assignment to a given population. Individual breed assignments to defined reference populations will be very difficult when introgression has occurred. The Bayesian approach within the Franches-Montagnes breed differentiated individuals with varied proportions of Warmblood.