Statistical Analysis of an SEIR Epidemic Model

This thesis was focussed on statistical analysis methods and proposes the use of Bayesian inference to extract information contained in experimental data by estimating Ebola model parameters. The model is a system of differential equations expressing the behavior and dynamics of Ebola. Two sets of data (onset and death data) were both used to estimate parameters, which has not been done by previous researchers in (Chowell, 2004). To be able to use both data, a new version of the model has been built. Model parameters have been estimated and then used to calculate the basic reproduction number and to study the disease-free equilibrium. Estimates of the parameters were useful to determine how well the model fits the data and how good estimates were, in terms of the information they provided about the possible relationship between variables. The solution showed that Ebola model fits the observed onset data at 98.95% and the observed death data at 93.6%. Since Bayesian inference can not be performed analytically, the Markov chain Monte Carlo approach has been used to generate samples from the posterior distribution over parameters. Samples have been used to check the accuracy of the model and other characteristics of the target posteriors.

Bayesian analysis of SEIR epidemic models

This thesis concerns the analysis of epidemic models. We adopt the Bayesian paradigm and develop suitable Markov Chain Monte Carlo (MCMC) algorithms. This is done by considering an Ebola outbreak in the Democratic Republic of Congo, former Zaïre, 1995 as a case of SEIR epidemic models. We model the Ebola epidemic deterministically using ODEs and stochastically through SDEs to take into account a possible bias in each compartment. Since the model has unknown parameters, we use different methods to estimate them such as least squares, maximum likelihood and MCMC. The motivation behind choosing MCMC over other existing methods in this thesis is that it has the ability to tackle complicated nonlinear problems with large number of parameters. First, in a deterministic Ebola model, we compute the likelihood function by sum of square of residuals method and estimate parameters using the LSQ and MCMC methods. We sample parameters and then use them to calculate the basic reproduction number and to study the disease-free equilibrium. From the sampled chain from the posterior, we test the convergence diagnostic and confirm the viability of the model. The results show that the Ebola model fits the observed onset data with high precision, and all the unknown model parameters are well identified. Second, we convert the ODE model into a SDE Ebola model. We compute the likelihood function using extended Kalman filter (EKF) and estimate parameters again. The motivation of using the SDE formulation here is to consider the impact of modelling errors. Moreover, the EKF approach allows us to formulate a filtered likelihood for the parameters of such a stochastic model. We use the MCMC procedure to attain the posterior distributions of the parameters of the SDE Ebola model drift and diffusion parts. In this thesis, we analyse two cases: (1) the model error covariance matrix of the dynamic noise is close to zero , i.e. only small stochasticity added into the model. The results are then similar to the ones got from deterministic Ebola model, even if methods of computing the likelihood function are different (2) the model error covariance matrix is different from zero, i.e. a considerable stochasticity is introduced into the Ebola model. This accounts for the situation where we would know that the model is not exact. As a results, we obtain parameter posteriors with larger variances. Consequently, the model predictions then show larger uncertainties, in accordance with the assumption of an incomplete model.

Naudan respiratory syncytial -virus ja naudan koronavirus-epidemiat keväällä ja tilanne syksyllä 2000

Summary: The spread of bovine respiratory syncytial virus and bovine coronavirus epidemic in spring and situation in fall 2000

Bordetella pertussis - Vaccination and Strain Variation

Pertussis or whooping cough is a highly contagious vaccine-preventable disease of the human respiratory tract caused by the Bordetella pertussis bacteria. In Finland, pertussis vaccinations were started in 1952 leading to a dramatic decrease in the morbidity and mortality. In the late 1990s, the incidence of pertussis increased despite the high vaccination coverage. Strain variation has been connected to the re-emergence of pertussis in countries with long history of pertussis vaccination. In 2005, the pertussis vaccine and the vaccination schedule were changed in Finland. The molecular epidemiology and the strain variation of the B. pertussis isolates were examined in Finland and in countries with similar (France) and different (Sweden) vaccination history. Continuous evolution of the B. pertussis population in Finland was observed since the 1950s, and the recently circulating isolates were antigenically different from the vaccine strains. Comparison of the circulating isolates from Finland, France and Sweden did not refer to significant differences. Certain type of strains noticed in France already in 1994 mainly caused the recent epidemics in Sweden (1999) and in Finland (2003-4). On several occasions, a new type of strains first appeared in Sweden and some years later in Finland. The B. pertussis isolates from the infants were shown to be similar to those from the other age groups. It is suggested that the strains originate from the same reservoir among adolescents and adults. The strain variation does not seem to have a major effect on the morbidity among recently vaccinated individuals, but it might play a role among those who are in the waning phase of immunity. The incidence of pertussis in Finland has remained low since the change of the vaccination programme. This might be related to the epidemic nature of pertussis and the near future will show the real effectiveness of the new vaccination programme. At present, many infants are infected because they are too young to be immunised with the current schedule. New strategies or vaccines are needed to protect those who are the most vulnerable.

Use of pyrosequencing to identify streptococci and to detect mutations causing antimicrobial resistance

Rapid identification and resistance determination of pathogens in clinical specimens is vital for accurate treatment and monitoring of infectious diseases. Antimicrobial drug resistance is increasing globally and healthcare settings are facing this cost-intensive and even life-threatening problem. The incidence of resistant pathogens in Finland has remained relatively steady and manageable at least for the time being. DNA sequencing is the gold standard method for genotyping, mutation analysis, and identification of bacteria. Due to significant cost decrease in recent years, this technique is available to many research and clinical laboratories. Pyrosequencing technique, a rapid real-time DNA sequencing method especially suitable for analyzing fairly short stretches of DNA, was used in this study. Due to its robustness and versatility, pyrosequencing was applied in this study for identification of streptococci and detection of certain mutations causing antimicrobial resistance in different bacteria. Certain streptococcal species such as S. pneumoniae and S. pyogenes are significantly important clinical pathogens. S. pneumoniae causes e.g. pneumonia and otitis media and is one of the most important community-acquired pathogens. S. pyogenes, also known as group A streptococcus, causes e.g. angina and erysipelas. In contrast, the socalled alpha-haemolytic streptococci, such as S. mitis and S. oralis, belong to the normal microbiota, which are regarded to be non-pathogenic and are nearly impossible to identify by phenotypic methods. In this thesis, a pyrosequencing method was developed for identification of streptococcal species based on the 16S rRNA sequences. Almost all streptococcal species could be differentiated from one another by the developed method, including S. pneumoniae from its close relatives S. mitis and S. oralis . New resistance genes and their variants are constantly discovered and reported. In this study, new methods for detecting certain mutations causing macrolide resistance or extended spectrum beta-lactamase (ESBL) phenotype were developed. These resistance detection approaches are not only suitable for surveillance of mechanisms causing antimicrobial resistance but also for routine analysis of clinical samples particularly in epidemic settings. In conclusion, pyrosequencing was found to be an accurate, versatile, cost-effective, and rapid DNA sequencing method that is especially suitable for mutation analysis of short DNA fragments and identification of certain bacteria.

Esikaupunkien ruoka-aavikoilta urbaaniin terveysajatteluun

Kirjallisuusarvostelu

Two-Photon Excited Fluorescence Detection Technology in Screening for Methicillin-Resistant Staphylococcus aureus

Control of the world-wide spread of methicillin-resistant Staphylococcus aureus (MRSA) has been unsuccessful in most developed countries. A few countries have been able to maintain a low MRSA prevalence, plausibly due to their strict MRSA control policies. Such policies require wide-scale screening of patients with suspected MRSA colonization, in order to nurse the MRSA-positive patients in contact isolation. The aim of this study was to develop and introduce a 2-photon excited fluorescence detection (TPX) technique for screening of MRSA directly from clinical samples. The assay principle involves specific online immunometric monitoring of S. aureus growth under selective antibiotic pressure. After the novel TPX approach had been set up, its applicability for the detection of MRSA was evaluated using a large MRSA collection including practically all epidemic MRSA strains identified in Finland between 1991 and 2009. The TPX assay was found both sensitive (97.9%) and specific (94.1%) in this epidemiological setting, illustrating that the method is tolerant to wide biological variation as well as to environments with rapidly emerging MRSA strains. When MRSA was screened directly from colonization samples, all patients positive for MRSA by conventional methods were positive also by the TPX assay. The assay capacity was 48 samples per a test run, and the median time required for confirmation of a true-positive screening test result was 3 h 26 min. Collectively, the findings presented in this thesis suggest that the TPX MRSA screening assay could be applicable for direct screening of MRSA colonization samples without any prior steps of isolation. This can potentially mean that contact isolation of suspected carriers testing negative could be discontinued earlier, thereby reducing the costs and burden associated with the containment of MRSA. In case of infection, a positive test result would ensure an early onset of effective therapy.

Numerical Simulation of Stochastic Di erential Equations

Stochastic differential equation (SDE) is a differential equation in which some of the terms and its solution are stochastic processes. SDEs play a central role in modeling physical systems like finance, Biology, Engineering, to mention some. In modeling process, the computation of the trajectories (sample paths) of solutions to SDEs is very important. However, the exact solution to a SDE is generally difficult to obtain due to non-differentiability character of realizations of the Brownian motion. There exist approximation methods of solutions of SDE. The solutions will be continuous stochastic processes that represent diffusive dynamics, a common modeling assumption for financial, Biology, physical, environmental systems. This Masters' thesis is an introduction and survey of numerical solution methods for stochastic differential equations. Standard numerical methods, local linearization methods and filtering methods are well described. We compute the root mean square errors for each method from which we propose a better numerical scheme. Stochastic differential equations can be formulated from a given ordinary differential equations. In this thesis, we describe two kind of formulations: parametric and non-parametric techniques. The formulation is based on epidemiological SEIR model. This methods have a tendency of increasing parameters in the constructed SDEs, hence, it requires more data. We compare the two techniques numerically.

From genotype to phenotype : diversity and population structure of Flavobacterium psychrophilum

Flavobacterium psychrophilum is the etiological agent of bacterial cold-water disease (BCWD) causing high fish mortalities and significant economic losses to the freshwater salmonid aquaculture industry around the world. Today BCWD outbreaks are mainly treated with environmentally hazardous antimicrobial agents and alternative preventative measures are urgently needed in order to ensure the well-being of animals and the sustainability of aquaculture. The diversity of pathogenic bacteria challenges the development of universal control strategies and in many cases the pathogen population structure, i.e. the total genetic diversity of the species must be taken into account. This work integrates the tools of modern molecular biology and conventional phenotypic microbiology to gain knowledge about the diversity and population structure of F. psychrophilum. The present work includes genetic characterization of a large collection of isolates collected from diverse origins and years, from aquaculture in a whole region including different countries, and provides the first international validation of a universal multilocus sequence typing (MLST) approach for unambiguous genetic typing of F. psychrophilum. Population structure analyses showed that the global F. psychrophilum population is subdivided into pathogenic species-specific clones, of which one particular genetic lineage, clonal complex CC-ST2, has been responsible for the majority of BCWD outbreaks in rainbow trout (Oncorhynchus mykiss) in European aquaculture facilities over several decades. Genotypic and phenotypic population heterogeneity affecting antimicrobial resistance in F. psychrophilum within BCWD outbreaks was discovered. Specific genotypes were associated with severe infections in farmed rainbow trout and Atlantic salmon (Salmo salar), and in addition to high adherence, antimicrobial resistance was strongly associated with outbreak strains. The study brought additional support for the hypothesis of an epidemic F. psychrophilum population structure, where recombination is an important force for the generation and maintenance of genetic diversity, and a significant contribution towards mapping the genetic diversity of this important fish pathogen. Evidence indicating dissemination of virulent strains with commercial movement of fish and fish products was strengthened.

Gut Microbiota and Metabolic Disorders

Obesity and its co-morbidities, such as metabolic syndrome (MetS), non-alcoholic fatty liver disease (NAFLD) and type 2 diabetes, have increased over the last few decades like an epidemic. So far the mechanisms of many metabolic diseases are not known in detail and currently there are not enough effective means to prevent and treat them. Several recent studies have shown that the unbalanced gut microbiota composition (GMC) and activity have an influence on the fat accumulation in the body. Further, it seems that the GMC of obese individuals differs from the lean. The aim of this study was to investigate whether there are differences between the GMC of metabolically impaired overweight/obese (MetS group), metabolically healthy overweight/obese and normal-weight individuals. In addition, the mechanisms by which the gut bacteria as well as their specific structures, such as flagellin (FLG) that stimulates the Toll-like receptor 5 (TLR5) affect metabolism, were investigated both in vivo and in vitro in human adipocytes and hepatocytes. The results of this study show that the abundance of certain gram-positive bacteria belonging to the Clostridial cluster XIV was higher in the MetS group subjects compared to their metabolically healthy overweight/obese and lean counterparts. Metabolically impaired subjects tended to also have a greater abundance of potentionally inflammatory Enterobacteria in their gut and thus seemed to have aberrant GMC. In addition, it was found that subjects with a high hepatic fat content (HHFC group) had less Faecalibacterium prausnitzii in their gut than individuals with low hepatic fat content. Further gene expression analysis revealed that the HHFC group also had increased inflammation cascades in their adipose tissue. Additionally, metabolically impaired individuals displayed an increased expression of FLG-recognizing TLR5 in adipose tissue, and the TLR5 expression levels associated positively both with liver fat content and insulin resistance in humans. These changes in the adipose tissue may further contribute to the impaired metabolism observed, such as insulin resistance and dyslipidemia. In vitro -studies showed that the FLG-induced TLR5 activation in adipocytes enhanced the hepatic fat accumulation by decreasing insulin signaling and mitochondrial functions and increasing triglyceride synthesis due to increased glycerol secretion from adipocytes. In conclusion, the findings of this study suggest that it may be possible that the novel prevention and personalized treatment strategies based on GM modulation will succesfully be developed for obesity and metabolic disorders in the future.