Bayesian Inference for Retrospective Population Genetics Models Using Markov Chain Monte Carlo Methods

Genetics, the science of heredity and variation in living organisms, has a central role in medicine, in breeding crops and livestock, and in studying fundamental topics of biological sciences such as evolution and cell functioning. Currently the field of genetics is under a rapid development because of the recent advances in technologies by which molecular data can be obtained from living organisms. In order that most information from such data can be extracted, the analyses need to be carried out using statistical models that are tailored to take account of the particular genetic processes. In this thesis we formulate and analyze Bayesian models for genetic marker data of contemporary individuals. The major focus is on the modeling of the unobserved recent ancestry of the sampled individuals (say, for tens of generations or so), which is carried out by using explicit probabilistic reconstructions of the pedigree structures accompanied by the gene flows at the marker loci. For such a recent history, the recombination process is the major genetic force that shapes the genomes of the individuals, and it is included in the model by assuming that the recombination fractions between the adjacent markers are known. The posterior distribution of the unobserved history of the individuals is studied conditionally on the observed marker data by using a Markov chain Monte Carlo algorithm (MCMC). The example analyses consider estimation of the population structure, relatedness structure (both at the level of whole genomes as well as at each marker separately), and haplotype configurations. For situations where the pedigree structure is partially known, an algorithm to create an initial state for the MCMC algorithm is given. Furthermore, the thesis includes an extension of the model for the recent genetic history to situations where also a quantitative phenotype has been measured from the contemporary individuals. In that case the goal is to identify positions on the genome that affect the observed phenotypic values. This task is carried out within the Bayesian framework, where the number and the relative effects of the quantitative trait loci are treated as random variables whose posterior distribution is studied conditionally on the observed genetic and phenotypic data. In addition, the thesis contains an extension of a widely-used haplotyping method, the PHASE algorithm, to settings where genetic material from several individuals has been pooled together, and the allele frequencies of each pool are determined in a single genotyping.

Monte Carlo Simulations of Molecular Clusters in Nucleation

A better understanding of the limiting step in a first order phase transition, the nucleation process, is of major importance to a variety of scientific fields ranging from atmospheric sciences to nanotechnology and even to cosmology. This is due to the fact that in most phase transitions the new phase is separated from the mother phase by a free energy barrier. This barrier is crossed in a process called nucleation. Nowadays it is considered that a significant fraction of all atmospheric particles is produced by vapor-to liquid nucleation. In atmospheric sciences, as well as in other scientific fields, the theoretical treatment of nucleation is mostly based on a theory known as the Classical Nucleation Theory. However, the Classical Nucleation Theory is known to have only a limited success in predicting the rate at which vapor-to-liquid nucleation takes place at given conditions. This thesis studies the unary homogeneous vapor-to-liquid nucleation from a statistical mechanics viewpoint. We apply Monte Carlo simulations of molecular clusters to calculate the free energy barrier separating the vapor and liquid phases and compare our results against the laboratory measurements and Classical Nucleation Theory predictions. According to our results, the work of adding a monomer to a cluster in equilibrium vapour is accurately described by the liquid drop model applied by the Classical Nucleation Theory, once the clusters are larger than some threshold size. The threshold cluster sizes contain only a few or some tens of molecules depending on the interaction potential and temperature. However, the error made in modeling the smallest of clusters as liquid drops results in an erroneous absolute value for the cluster work of formation throughout the size range, as predicted by the McGraw-Laaksonen scaling law. By calculating correction factors to Classical Nucleation Theory predictions for the nucleation barriers of argon and water, we show that the corrected predictions produce nucleation rates that are in good comparison with experiments. For the smallest clusters, the deviation between the simulation results and the liquid drop values are accurately modelled by the low order virial coefficients at modest temperatures and vapour densities, or in other words, in the validity range of the non-interacting cluster theory by Frenkel, Band and Bilj. Our results do not indicate a need for a size dependent replacement free energy correction. The results also indicate that Classical Nucleation Theory predicts the size of the critical cluster correctly. We also presents a new method for the calculation of the equilibrium vapour density, surface tension size dependence and planar surface tension directly from cluster simulations. We also show how the size dependence of the cluster surface tension in equimolar surface is a function of virial coefficients, a result confirmed by our cluster simulations.

Monte Carlo simulations of molecular clusters in nucleation

A better understanding of the limiting step in a first order phase transition, the nucleation process, is of major importance to a variety of scientific fields ranging from atmospheric sciences to nanotechnology and even to cosmology. This is due to the fact that in most phase transitions the new phase is separated from the mother phase by a free energy barrier. This barrier is crossed in a process called nucleation. Nowadays it is considered that a significant fraction of all atmospheric particles is produced by vapor-to liquid nucleation. In atmospheric sciences, as well as in other scientific fields, the theoretical treatment of nucleation is mostly based on a theory known as the Classical Nucleation Theory. However, the Classical Nucleation Theory is known to have only a limited success in predicting the rate at which vapor-to-liquid nucleation takes place at given conditions. This thesis studies the unary homogeneous vapor-to-liquid nucleation from a statistical mechanics viewpoint. We apply Monte Carlo simulations of molecular clusters to calculate the free energy barrier separating the vapor and liquid phases and compare our results against the laboratory measurements and Classical Nucleation Theory predictions. According to our results, the work of adding a monomer to a cluster in equilibrium vapour is accurately described by the liquid drop model applied by the Classical Nucleation Theory, once the clusters are larger than some threshold size. The threshold cluster sizes contain only a few or some tens of molecules depending on the interaction potential and temperature. However, the error made in modeling the smallest of clusters as liquid drops results in an erroneous absolute value for the cluster work of formation throughout the size range, as predicted by the McGraw-Laaksonen scaling law. By calculating correction factors to Classical Nucleation Theory predictions for the nucleation barriers of argon and water, we show that the corrected predictions produce nucleation rates that are in good comparison with experiments. For the smallest clusters, the deviation between the simulation results and the liquid drop values are accurately modelled by the low order virial coefficients at modest temperatures and vapour densities, or in other words, in the validity range of the non-interacting cluster theory by Frenkel, Band and Bilj. Our results do not indicate a need for a size dependent replacement free energy correction. The results also indicate that Classical Nucleation Theory predicts the size of the critical cluster correctly. We also presents a new method for the calculation of the equilibrium vapour density, surface tension size dependence and planar surface tension directly from cluster simulations. We also show how the size dependence of the cluster surface tension in equimolar surface is a function of virial coefficients, a result confirmed by our cluster simulations.

Bootstrapping the Error Correction Model Cointegration Test

This paper is concerned with using the bootstrap to obtain improved critical values for the error correction model (ECM) cointegration test in dynamic models. In the paper we investigate the effects of dynamic specification on the size and power of the ECM cointegration test with bootstrap critical values. The results from a Monte Carlo study show that the size of the bootstrap ECM cointegration test is close to the nominal significance level. We find that overspecification of the lag length results in a loss of power. Underspecification of the lag length results in size distortion. The performance of the bootstrap ECM cointegration test deteriorates if the correct lag length is not used in the ECM. The bootstrap ECM cointegration test is therefore not robust to model misspecification.

Conceptual and statistical modelling of environmental effects in population dynamics

Population dynamics are generally viewed as the result of intrinsic (purely density dependent) and extrinsic (environmental) processes. Both components, and potential interactions between those two, have to be modelled in order to understand and predict dynamics of natural populations; a topic that is of great importance in population management and conservation. This thesis focuses on modelling environmental effects in population dynamics and how effects of potentially relevant environmental variables can be statistically identified and quantified from time series data. Chapter I presents some useful models of multiplicative environmental effects for unstructured density dependent populations. The presented models can be written as standard multiple regression models that are easy to fit to data. Chapters II IV constitute empirical studies that statistically model environmental effects on population dynamics of several migratory bird species with different life history characteristics and migration strategies. In Chapter II, spruce cone crops are found to have a strong positive effect on the population growth of the great spotted woodpecker (Dendrocopos major), while cone crops of pine another important food resource for the species do not effectively explain population growth. The study compares rate- and ratio-dependent effects of cone availability, using state-space models that distinguish between process and observation error in the time series data. Chapter III shows how drought, in combination with settling behaviour during migration, produces asymmetric spatially synchronous patterns of population dynamics in North American ducks (genus Anas). Chapter IV investigates the dynamics of a Finnish population of skylark (Alauda arvensis), and point out effects of rainfall and habitat quality on population growth. Because the skylark time series and some of the environmental variables included show strong positive autocorrelation, the statistical significances are calculated using a Monte Carlo method, where random autocorrelated time series are generated. Chapter V is a simulation-based study, showing that ignoring observation error in analyses of population time series data can bias the estimated effects and measures of uncertainty, if the environmental variables are autocorrelated. It is concluded that the use of state-space models is an effective way to reach more accurate results. In summary, there are several biological assumptions and methodological issues that can affect the inferential outcome when estimating environmental effects from time series data, and that therefore need special attention. The functional form of the environmental effects and potential interactions between environment and population density are important to deal with. Other issues that should be considered are assumptions about density dependent regulation, modelling potential observation error, and when needed, accounting for spatial and/or temporal autocorrelation.

Auger-emitter radiochemotherapy in squamous cell cancers : in vitro and in vivo experiments with In-111-BLMC

Head and neck squamous cell cancer (HNSCC) is the sixth most common cancer worldwide. Despite advances in combined modality therapy (surgery, radiotherapy, chemotherapy) the 5-year survival rate in stage III and IV disease remains at 40% - 60%. Short-range Auger-electron emitters, such as In-111 and In-114m, tagged with a drug, molecule, peptide, protein or nanoparticles brought in close proximity to nuclear DNA represent a fascinating alternative for treating cancer. In this thesis, we studied the usefulness of Indium-111-bleomycin complex (In-111-BLMC) in the diagnostics and potential therapy of HNSCC using in vitro HNSCC cell lines, in vivo nude mice, and in vivo HNSCC patients. In in vitro experiments with HNSCC cell lines, the sensitivity to external beam radiation, BLM, In-111-BLMC, and In-111-Cl3 was studied using the 96-well plate clonogenic assay. The influence of BLM and In-111-BLMC on the cell cycle was measured with flow cytometry. In in vivo nude mice xenograft studies, the activity ratios of In-111-BLMC were obtained in gamma camera images. The effect of In-111-BLMC in HNSCC xenografts was studied. In in vivo patient studies, we determined the tumor uptake of In-111-BLMC with gamma camera and the radioactivity from tumor samples using In-111-BLMC with specific activity of 75, 175, or 375 MBq/mg BLM. The S values, i.e. absorbed dose in a target organ per cumulated activity in a source organ, were simulated for In-111 and In-114m. In vitro studies showed the variation of sensitivity for external beam radiation, BLM, and In-111-BLMC between HNSCC cell lines. IC50 values for BLM were 1.6-, 1.8-, and 2.1-fold higher than In-111-BLMC (40 MBq/mg BLM) in three HNSCC cell lines. Specific In-111 activity of 40 MBq/mgBLM was more effective in killing cells than specific In-111 activity of 195MBq/mgBLM (p=0.0023). In-111-Cl3 alone had no killing effect. The percentage of cells in the G2/M phase increased after exposure to BLM and especially to In-111-BLMC in the three cell lines studied, indicating a G2/M block. The tumor-seeking behavior was shown in the in vivo imaging study of xenografted mice. BLM and In-111-BLMC were more effective than NaCl in reducing xenografted tumor size in HNSCC. The uptake ratios received from gamma images in the in vivo patient study varied from 1.2 to 2.8 in malignant tumors. However, the uptake of In-111-BLMC was unaffected by increasing the injected activity. A positive correlation existed between In-111-BLMC uptake, Ki-67/MIB activity, and number of mitoses. Regarding the S values, In-114m delivered a 4-fold absorbed radiation dose into the tumor compared with In-111, and thus, In-114m-BLMC might be more effective than In-111-BLMC at the DNA level. Auger-electron emitters, such as In-111 and In-114m, might have potential in the treatment of HNSCC. Further studies are needed to develop a radiopharmaceutical agent with appropriate physical properties of the radionuclide and a suitable carrier to bring it to the targeted tissue.

Optical Studies of the Antarctic Glacio-Oceanic System

The Antarctic system comprises of the continent itself, Antarctica, and the ocean surrounding it, the Southern Ocean. The system has an important part in the global climate due to its size, its high latitude location and the negative radiation balance of its large ice sheets. Antarctica has also been in focus for several decades due to increased ultraviolet (UV) levels caused by stratospheric ozone depletion, and the disintegration of its ice shelves. In this study, measurements were made during three Austral summers to study the optical properties of the Antarctic system and to produce radiation information for additional modeling studies. These are related to specific phenomena found in the system. During the summer of 1997-1998, measurements of beam absorption and beam attenuation coefficients, and downwelling and upwelling irradiance were made in the Southern Ocean along a S-N transect at 6°E. The attenuation of photosynthetically active radiation (PAR) was calculated and used together with hydrographic measurements to judge whether the phytoplankton in the investigated areas of the Southern Ocean are light limited. By using the Kirk formula the diffuse attenuation coefficient was linked to the absorption and scattering coefficients. The diffuse attenuation coefficients (Kpar) for PAR were found to vary between 0.03 and 0.09 1/m. Using the values for KPAR and the definition of the Sverdrup critical depth, the studied Southern Ocean plankton systems were found not to be light limited. Variabilities in the spectral and total albedo of snow were studied in the Queen Maud Land region of Antarctica during the summers of 1999-2000 and 2000-2001. The measurement areas were the vicinity of the South African Antarctic research station SANAE 4, and a traverse near the Finnish Antarctic research station Aboa. The midday mean total albedos for snow were between 0.83, for clear skies, and 0.86, for overcast skies, at Aboa and between 0.81 and 0.83 for SANAE 4. The mean spectral albedo levels at Aboa and SANAE 4 were very close to each other. The variations in the spectral albedos were due more to differences in ambient conditions than variations in snow properties. A Monte-Carlo model was developed to study the spectral albedo and to develop a novel nondestructive method to measure the diffuse attenuation coefficient of snow. The method was based on the decay of upwelling radiation moving horizontally away from a source of downwelling light. This was assumed to have a relation to the diffuse attenuation coefficient. In the model, the attenuation coefficient obtained from the upwelling irradiance was higher than that obtained using vertical profiles of downwelling irradiance. The model results were compared to field measurements made on dry snow in Finnish Lapland and they correlated reasonably well. Low-elevation (below 1000 m) blue-ice areas may experience substantial melt-freeze cycles due to absorbed solar radiation and the small heat conductivity in the ice. A two-dimensional (x-z) model has been developed to simulate the formation and water circulation in the subsurface ponds. The model results show that for a physically reasonable parameter set the formation of liquid water within the ice can be reproduced. The results however are sensitive to the chosen parameter values, and their exact values are not well known. Vertical convection and a weak overturning circulation is generated stratifying the fluid and transporting warmer water downward, thereby causing additional melting at the base of the pond. In a 50-year integration, a global warming scenario mimicked by a decadal scale increase of 3 degrees per 100 years in air temperature, leads to a general increase in subsurface water volume. The ice did not disintegrate due to the air temperature increase after the 50 year integration.

Irradiation-mediated tailoring of carbon nanotubes

The ever-increasing demand for faster computers in various areas, ranging from entertaining electronics to computational science, is pushing the semiconductor industry towards its limits on decreasing the sizes of electronic devices based on conventional materials. According to the famous law by Gordon E. Moore, a co-founder of the world s largest semiconductor company Intel, the transistor sizes should decrease to the atomic level during the next few decades to maintain the present rate of increase in the computational power. As leakage currents become a problem for traditional silicon-based devices already at sizes in the nanometer scale, an approach other than further miniaturization is needed to accomplish the needs of the future electronics. A relatively recently proposed possibility for further progress in electronics is to replace silicon with carbon, another element from the same group in the periodic table. Carbon is an especially interesting material for nanometer-sized devices because it forms naturally different nanostructures. Furthermore, some of these structures have unique properties. The most widely suggested allotrope of carbon to be used for electronics is a tubular molecule having an atomic structure resembling that of graphite. These carbon nanotubes are popular both among scientists and in industry because of a wide list of exciting properties. For example, carbon nanotubes are electronically unique and have uncommonly high strength versus mass ratio, which have resulted in a multitude of proposed applications in several fields. In fact, due to some remaining difficulties regarding large-scale production of nanotube-based electronic devices, fields other than electronics have been faster to develop profitable nanotube applications. In this thesis, the possibility of using low-energy ion irradiation to ease the route towards nanotube applications is studied through atomistic simulations on different levels of theory. Specifically, molecular dynamic simulations with analytical interaction models are used to follow the irradiation process of nanotubes to introduce different impurity atoms into these structures, in order to gain control on their electronic character. Ion irradiation is shown to be a very efficient method to replace carbon atoms with boron or nitrogen impurities in single-walled nanotubes. Furthermore, potassium irradiation of multi-walled and fullerene-filled nanotubes is demonstrated to result in small potassium clusters in the hollow parts of these structures. Molecular dynamic simulations are further used to give an example on using irradiation to improve contacts between a nanotube and a silicon substrate. Methods based on the density-functional theory are used to gain insight on the defect structures inevitably created during the irradiation. Finally, a new simulation code utilizing the kinetic Monte Carlo method is introduced to follow the time evolution of irradiation-induced defects on carbon nanotubes on macroscopic time scales. Overall, the molecular dynamic simulations presented in this thesis show that ion irradiation is a promisingmethod for tailoring the nanotube properties in a controlled manner. The calculations made with density-functional-theory based methods indicate that it is energetically favorable for even relatively large defects to transform to keep the atomic configuration as close to the pristine nanotube as possible. The kinetic Monte Carlo studies reveal that elevated temperatures during the processing enhance the self-healing of nanotubes significantly, ensuring low defect concentrations after the treatment with energetic ions. Thereby, nanotubes can retain their desired properties also after the irradiation. Throughout the thesis, atomistic simulations combining different levels of theory are demonstrated to be an important tool for determining the optimal conditions for irradiation experiments, because the atomic-scale processes at short time scales are extremely difficult to study by any other means.

Generalised regression estimation for domain class frequencies

This study examines the properties of Generalised Regression (GREG) estimators for domain class frequencies and proportions. The family of GREG estimators forms the class of design-based model-assisted estimators. All GREG estimators utilise auxiliary information via modelling. The classic GREG estimator with a linear fixed effects assisting model (GREG-lin) is one example. But when estimating class frequencies, the study variable is binary or polytomous. Therefore logistic-type assisting models (e.g. logistic or probit model) should be preferred over the linear one. However, other GREG estimators than GREG-lin are rarely used, and knowledge about their properties is limited. This study examines the properties of L-GREG estimators, which are GREG estimators with fixed-effects logistic-type models. Three research questions are addressed. First, I study whether and when L-GREG estimators are more accurate than GREG-lin. Theoretical results and Monte Carlo experiments which cover both equal and unequal probability sampling designs and a wide variety of model formulations show that in standard situations, the difference between L-GREG and GREG-lin is small. But in the case of a strong assisting model, two interesting situations arise: if the domain sample size is reasonably large, L-GREG is more accurate than GREG-lin, and if the domain sample size is very small, estimation of assisting model parameters may be inaccurate, resulting in bias for L-GREG. Second, I study variance estimation for the L-GREG estimators. The standard variance estimator (S) for all GREG estimators resembles the Sen-Yates-Grundy variance estimator, but it is a double sum of prediction errors, not of the observed values of the study variable. Monte Carlo experiments show that S underestimates the variance of L-GREG especially if the domain sample size is minor, or if the assisting model is strong. Third, since the standard variance estimator S often fails for the L-GREG estimators, I propose a new augmented variance estimator (A). The difference between S and the new estimator A is that the latter takes into account the difference between the sample fit model and the census fit model. In Monte Carlo experiments, the new estimator A outperformed the standard estimator S in terms of bias, root mean square error and coverage rate. Thus the new estimator provides a good alternative to the standard estimator.

Measurement of Particle Production and Inclusive Differential Cross Sections in pbar{p} Collisions at sqrt{s}=1.96 TeV

We report a set of measurements of particle production in inelastic pbar{p} collisions collected with a minimum-bias trigger at the Tevatron Collider with the CDF II experiment. The inclusive charged particle transverse momentum differential cross section is measured, with improved precision, over a range about ten times wider than in previous measurements. The former modeling of the spectrum appears to be incompatible with the high particle momenta observed. The dependence of the charged particle transverse momentum on the event particle multiplicity is analyzed to study the various components of hadron interactions. This is one of the observable variables most poorly reproduced by the available Monte Carlo generators. A first measurement of the event transverse energy sum differential cross section is also reported. A comparison with a Pythia prediction at the hadron level is performed. The inclusive charged particle differential production cross section is fairly well reproduced only in the transverse momentum range available from previous measurements. At higher momentum the agreement is poor. The transverse energy sum is poorly reproduced over the whole spectrum. The dependence of the charged particle transverse momentum on the particle multiplicity needs the introduction of more sophisticated particle production mechanisms, such as multiple parton interactions, in order to be better explained.

Measurement of particle production and inclusive differential cross sections in pp̅ collisions at √s=1.96 TeV

We report a set of measurements of particle production in inelastic pbar{p} collisions collected with a minimum-bias trigger at the Tevatron Collider with the CDF II experiment. The inclusive charged particle transverse momentum differential cross section is measured, with improved precision, over a range about ten times wider than in previous measurements. The former modeling of the spectrum appears to be incompatible with the high particle momenta observed. The dependence of the charged particle transverse momentum on the event particle multiplicity is analyzed to study the various components of hadron interactions. This is one of the observable variables most poorly reproduced by the available Monte Carlo generators. A first measurement of the event transverse energy sum differential cross section is also reported. A comparison with a Pythia prediction at the hadron level is performed. The inclusive charged particle differential production cross section is fairly well reproduced only in the transverse momentum range available from previous measurements. At higher momentum the agreement is poor. The transverse energy sum is poorly reproduced over the whole spectrum. The dependence of the charged particle transverse momentum on the particle multiplicity needs the introduction of more sophisticated particle production mechanisms, such as multiple parton interactions, in order to be better explained.

Markov random fields and spatial smoothing in improving of forest inventory estimates

Markov random fields (MRF) are popular in image processing applications to describe spatial dependencies between image units. Here, we take a look at the theory and the models of MRFs with an application to improve forest inventory estimates. Typically, autocorrelation between study units is a nuisance in statistical inference, but we take an advantage of the dependencies to smooth noisy measurements by borrowing information from the neighbouring units. We build a stochastic spatial model, which we estimate with a Markov chain Monte Carlo simulation method. The smooth values are validated against another data set increasing our confidence that the estimates are more accurate than the originals.