Computational Methods for Detecting Large-Scale Chromosome Rearrangements in SNP Data

Large-scale chromosome rearrangements such as copy number variants (CNVs) and inversions encompass a considerable proportion of the genetic variation between human individuals. In a number of cases, they have been closely linked with various inheritable diseases. Single-nucleotide polymorphisms (SNPs) are another large part of the genetic variance between individuals. They are also typically abundant and their measuring is straightforward and cheap. This thesis presents computational means of using SNPs to detect the presence of inversions and deletions, a particular variety of CNVs. Technically, the inversion-detection algorithm detects the suppressed recombination rate between inverted and non-inverted haplotype populations whereas the deletion-detection algorithm uses the EM-algorithm to estimate the haplotype frequencies of a window with and without a deletion haplotype. As a contribution to population biology, a coalescent simulator for simulating inversion polymorphisms has been developed. Coalescent simulation is a backward-in-time method of modelling population ancestry. Technically, the simulator also models multiple crossovers by using the Counting model as the chiasma interference model. Finally, this thesis includes an experimental section. The aforementioned methods were tested on synthetic data to evaluate their power and specificity. They were also applied to the HapMap Phase II and Phase III data sets, yielding a number of candidates for previously unknown inversions, deletions and also correctly detecting known such rearrangements.

Computational Techniques for Haplotype Inference and for Local Alignment Significance

This thesis which consists of an introduction and four peer-reviewed original publications studies the problems of haplotype inference (haplotyping) and local alignment significance. The problems studied here belong to the broad area of bioinformatics and computational biology. The presented solutions are computationally fast and accurate, which makes them practical in high-throughput sequence data analysis. Haplotype inference is a computational problem where the goal is to estimate haplotypes from a sample of genotypes as accurately as possible. This problem is important as the direct measurement of haplotypes is difficult, whereas the genotypes are easier to quantify. Haplotypes are the key-players when studying for example the genetic causes of diseases. In this thesis, three methods are presented for the haplotype inference problem referred to as HaploParser, HIT, and BACH. HaploParser is based on a combinatorial mosaic model and hierarchical parsing that together mimic recombinations and point-mutations in a biologically plausible way. In this mosaic model, the current population is assumed to be evolved from a small founder population. Thus, the haplotypes of the current population are recombinations of the (implicit) founder haplotypes with some point--mutations. HIT (Haplotype Inference Technique) uses a hidden Markov model for haplotypes and efficient algorithms are presented to learn this model from genotype data. The model structure of HIT is analogous to the mosaic model of HaploParser with founder haplotypes. Therefore, it can be seen as a probabilistic model of recombinations and point-mutations. BACH (Bayesian Context-based Haplotyping) utilizes a context tree weighting algorithm to efficiently sum over all variable-length Markov chains to evaluate the posterior probability of a haplotype configuration. Algorithms are presented that find haplotype configurations with high posterior probability. BACH is the most accurate method presented in this thesis and has comparable performance to the best available software for haplotype inference. Local alignment significance is a computational problem where one is interested in whether the local similarities in two sequences are due to the fact that the sequences are related or just by chance. Similarity of sequences is measured by their best local alignment score and from that, a p-value is computed. This p-value is the probability of picking two sequences from the null model that have as good or better best local alignment score. Local alignment significance is used routinely for example in homology searches. In this thesis, a general framework is sketched that allows one to compute a tight upper bound for the p-value of a local pairwise alignment score. Unlike the previous methods, the presented framework is not affeced by so-called edge-effects and can handle gaps (deletions and insertions) without troublesome sampling and curve fitting.

Computational Methods for Locating and Analyzing Conserved Gene Regulatory DNA Elements

This thesis presents methods for locating and analyzing cis-regulatory DNA elements involved with the regulation of gene expression in multicellular organisms. The regulation of gene expression is carried out by the combined effort of several transcription factor proteins collectively binding the DNA on the cis-regulatory elements. Only sparse knowledge of the 'genetic code' of these elements exists today. An automatic tool for discovery of putative cis-regulatory elements could help their experimental analysis, which would result in a more detailed view of the cis-regulatory element structure and function. We have developed a computational model for the evolutionary conservation of cis-regulatory elements. The elements are modeled as evolutionarily conserved clusters of sequence-specific transcription factor binding sites. We give an efficient dynamic programming algorithm that locates the putative cis-regulatory elements and scores them according to the conservation model. A notable proportion of the high-scoring DNA sequences show transcriptional enhancer activity in transgenic mouse embryos. The conservation model includes four parameters whose optimal values are estimated with simulated annealing. With good parameter values the model discriminates well between the DNA sequences with evolutionarily conserved cis-regulatory elements and the DNA sequences that have evolved neutrally. In further inquiry, the set of highest scoring putative cis-regulatory elements were found to be sensitive to small variations in the parameter values. The statistical significance of the putative cis-regulatory elements is estimated with the Two Component Extreme Value Distribution. The p-values grade the conservation of the cis-regulatory elements above the neutral expectation. The parameter values for the distribution are estimated by simulating the neutral DNA evolution. The conservation of the transcription factor binding sites can be used in the upstream analysis of regulatory interactions. This approach may provide mechanistic insight to the transcription level data from, e.g., microarray experiments. Here we give a method to predict shared transcriptional regulators for a set of co-expressed genes. The EEL (Enhancer Element Locator) software implements the method for locating putative cis-regulatory elements. The software facilitates both interactive use and distributed batch processing. We have used it to analyze the non-coding regions around all human genes with respect to the orthologous regions in various other species including mouse. The data from these genome-wide analyzes is stored in a relational database which is used in the publicly available web services for upstream analysis and visualization of the putative cis-regulatory elements in the human genome.

Modeling Efficient Classification as a Process of Confidence Assessment and Delegation

In visual object detection and recognition, classifiers have two interesting characteristics: accuracy and speed. Accuracy depends on the complexity of the image features and classifier decision surfaces. Speed depends on the hardware and the computational effort required to use the features and decision surfaces. When attempts to increase accuracy lead to increases in complexity and effort, it is necessary to ask how much are we willing to pay for increased accuracy. For example, if increased computational effort implies quickly diminishing returns in accuracy, then those designing inexpensive surveillance applications cannot aim for maximum accuracy at any cost. It becomes necessary to find trade-offs between accuracy and effort. We study efficient classification of images depicting real-world objects and scenes. Classification is efficient when a classifier can be controlled so that the desired trade-off between accuracy and effort (speed) is achieved and unnecessary computations are avoided on a per input basis. A framework is proposed for understanding and modeling efficient classification of images. Classification is modeled as a tree-like process. In designing the framework, it is important to recognize what is essential and to avoid structures that are narrow in applicability. Earlier frameworks are lacking in this regard. The overall contribution is two-fold. First, the framework is presented, subjected to experiments, and shown to be satisfactory. Second, certain unconventional approaches are experimented with. This allows the separation of the essential from the conventional. To determine if the framework is satisfactory, three categories of questions are identified: trade-off optimization, classifier tree organization, and rules for delegation and confidence modeling. Questions and problems related to each category are addressed and empirical results are presented. For example, related to trade-off optimization, we address the problem of computational bottlenecks that limit the range of trade-offs. We also ask if accuracy versus effort trade-offs can be controlled after training. For another example, regarding classifier tree organization, we first consider the task of organizing a tree in a problem-specific manner. We then ask if problem-specific organization is necessary.

Learning Nonlinear Visual Processing from Natural Images

The paradigm of computational vision hypothesizes that any visual function -- such as the recognition of your grandparent -- can be replicated by computational processing of the visual input. What are these computations that the brain performs? What should or could they be? Working on the latter question, this dissertation takes the statistical approach, where the suitable computations are attempted to be learned from the natural visual data itself. In particular, we empirically study the computational processing that emerges from the statistical properties of the visual world and the constraints and objectives specified for the learning process. This thesis consists of an introduction and 7 peer-reviewed publications, where the purpose of the introduction is to illustrate the area of study to a reader who is not familiar with computational vision research. In the scope of the introduction, we will briefly overview the primary challenges to visual processing, as well as recall some of the current opinions on visual processing in the early visual systems of animals. Next, we describe the methodology we have used in our research, and discuss the presented results. We have included some additional remarks, speculations and conclusions to this discussion that were not featured in the original publications. We present the following results in the publications of this thesis. First, we empirically demonstrate that luminance and contrast are strongly dependent in natural images, contradicting previous theories suggesting that luminance and contrast were processed separately in natural systems due to their independence in the visual data. Second, we show that simple cell -like receptive fields of the primary visual cortex can be learned in the nonlinear contrast domain by maximization of independence. Further, we provide first-time reports of the emergence of conjunctive (corner-detecting) and subtractive (opponent orientation) processing due to nonlinear projection pursuit with simple objective functions related to sparseness and response energy optimization. Then, we show that attempting to extract independent components of nonlinear histogram statistics of a biologically plausible representation leads to projection directions that appear to differentiate between visual contexts. Such processing might be applicable for priming, \ie the selection and tuning of later visual processing. We continue by showing that a different kind of thresholded low-frequency priming can be learned and used to make object detection faster with little loss in accuracy. Finally, we show that in a computational object detection setting, nonlinearly gain-controlled visual features of medium complexity can be acquired sequentially as images are encountered and discarded. We present two online algorithms to perform this feature selection, and propose the idea that for artificial systems, some processing mechanisms could be selectable from the environment without optimizing the mechanisms themselves. In summary, this thesis explores learning visual processing on several levels. The learning can be understood as interplay of input data, model structures, learning objectives, and estimation algorithms. The presented work adds to the growing body of evidence showing that statistical methods can be used to acquire intuitively meaningful visual processing mechanisms. The work also presents some predictions and ideas regarding biological visual processing.

Computational Identification of Recessive Mutations in Cancers using High Throughput SNP-arrays

This thesis presents a highly sensitive genome wide search method for recessive mutations. The method is suitable for distantly related samples that are divided into phenotype positives and negatives. High throughput genotype arrays are used to identify and compare homozygous regions between the cohorts. The method is demonstrated by comparing colorectal cancer patients against unaffected references. The objective is to find homozygous regions and alleles that are more common in cancer patients. We have designed and implemented software tools to automate the data analysis from genotypes to lists of candidate genes and to their properties. The programs have been designed in respect to a pipeline architecture that allows their integration to other programs such as biological databases and copy number analysis tools. The integration of the tools is crucial as the genome wide analysis of the cohort differences produces many candidate regions not related to the studied phenotype. CohortComparator is a genotype comparison tool that detects homozygous regions and compares their loci and allele constitutions between two sets of samples. The data is visualised in chromosome specific graphs illustrating the homozygous regions and alleles of each sample. The genomic regions that may harbour recessive mutations are emphasised with different colours and a scoring scheme is given for these regions. The detection of homozygous regions, cohort comparisons and result annotations are all subjected to presumptions many of which have been parameterized in our programs. The effect of these parameters and the suitable scope of the methods have been evaluated. Samples with different resolutions can be balanced with the genotype estimates of their haplotypes and they can be used within the same study.

Computational fluid dynamics simulations in aerosol and nucleation studies

Nucleation is the first step in the formation of a new phase inside a mother phase. Two main forms of nucleation can be distinguished. In homogeneous nucleation, the new phase is formed in a uniform substance. In heterogeneous nucleation, on the other hand, the new phase emerges on a pre-existing surface (nucleation site). Nucleation is the source of about 30% of all atmospheric aerosol which in turn has noticeable health effects and a significant impact on climate. Nucleation can be observed in the atmosphere, studied experimentally in the laboratory and is the subject of ongoing theoretical research. This thesis attempts to be a link between experiment and theory. By comparing simulation results to experimental data, the aim is to (i) better understand the experiments and (ii) determine where the theory needs improvement. Computational fluid dynamics (CFD) tools were used to simulate homogeneous onecomponent nucleation of n-alcohols in argon and helium as carrier gases, homogeneous nucleation in the water-sulfuric acid-system, and heterogeneous nucleation of water vapor on silver particles. In the nucleation of n-alcohols, vapor depletion, carrier gas effect and carrier gas pressure effect were evaluated, with a special focus on the pressure effect whose dependence on vapor and carrier gas properties could be specified. The investigation of nucleation in the water-sulfuric acid-system included a thorough analysis of the experimental setup, determining flow conditions, vapor losses, and nucleation zone. Experimental nucleation rates were compared to various theoretical approaches. We found that none of the considered theoretical descriptions of nucleation captured the role of water in the process at all relative humidities. Heterogeneous nucleation was studied in the activation of silver particles in a TSI 3785 particle counter which uses water as its working fluid. The role of the contact angle was investigated and the influence of incoming particle concentrations and homogeneous nucleation on counting efficiency determined.

Geant4 Hadronic Cascade Models and CMS Data Analysis : Computational Challenges in the LHC era

This work belongs to the field of computational high-energy physics (HEP). The key methods used in this thesis work to meet the challenges raised by the Large Hadron Collider (LHC) era experiments are object-orientation with software engineering, Monte Carlo simulation, the computer technology of clusters, and artificial neural networks. The first aspect discussed is the development of hadronic cascade models, used for the accurate simulation of medium-energy hadron-nucleus reactions, up to 10 GeV. These models are typically needed in hadronic calorimeter studies and in the estimation of radiation backgrounds. Various applications outside HEP include the medical field (such as hadron treatment simulations), space science (satellite shielding), and nuclear physics (spallation studies). Validation results are presented for several significant improvements released in Geant4 simulation tool, and the significance of the new models for computing in the Large Hadron Collider era is estimated. In particular, we estimate the ability of the Bertini cascade to simulate Compact Muon Solenoid (CMS) hadron calorimeter HCAL. LHC test beam activity has a tightly coupled cycle of simulation-to-data analysis. Typically, a Geant4 computer experiment is used to understand test beam measurements. Thus an another aspect of this thesis is a description of studies related to developing new CMS H2 test beam data analysis tools and performing data analysis on the basis of CMS Monte Carlo events. These events have been simulated in detail using Geant4 physics models, full CMS detector description, and event reconstruction. Using the ROOT data analysis framework we have developed an offline ANN-based approach to tag b-jets associated with heavy neutral Higgs particles, and we show that this kind of NN methodology can be successfully used to separate the Higgs signal from the background in the CMS experiment.

Theoretical and computational approaches on heterogeneous nucleation

Nucleation is the first step of a first order phase transition. A new phase is always sprung up in nucleation phenomena. The two main categories of nucleation are homogeneous nucleation, where the new phase is formed in a uniform substance, and heterogeneous nucleation, when nucleation occurs on a pre-existing surface. In this thesis the main attention is paid on heterogeneous nucleation. This thesis wields the nucleation phenomena from two theoretical perspectives: the classical nucleation theory and the statistical mechanical approach. The formulation of the classical nucleation theory relies on equilibrium thermodynamics and use of macroscopically determined quantities to describe the properties of small nuclei, sometimes consisting of just a few molecules. The statistical mechanical approach is based on interactions between single molecules, and does not bear the same assumptions as the classical theory. This work gathers up the present theoretical knowledge of heterogeneous nucleation and utilizes it in computational model studies. A new exact molecular approach on heterogeneous nucleation was introduced and tested by Monte Carlo simulations. The results obtained from the molecular simulations were interpreted by means of the concepts of the classical nucleation theory. Numerical calculations were carried out for a variety of substances nucleating on different substances. The classical theory of heterogeneous nucleation was employed in calculations of one-component nucleation of water on newsprint paper, Teflon and cellulose film, and binary nucleation of water-n-propanol and water-sulphuric acid mixtures on silver nanoparticles. The results were compared with experimental results. The molecular simulation studies involved homogeneous nucleation of argon and heterogeneous nucleation of argon on a planar platinum surface. It was found out that the use of a microscopical contact angle as a fitting parameter in calculations based on the classical theory of heterogeneous nucleation leads to a fair agreement between the theoretical predictions and experimental results. In the presented cases the microscopical angle was found to be always smaller than the contact angle obtained from macroscopical measurements. Furthermore, molecular Monte Carlo simulations revealed that the concept of the geometrical contact parameter in heterogeneous nucleation calculations can work surprisingly well even for very small clusters.

New computational approaches for inelastic x-ray scattering

Inelastic x-ray scattering spectroscopy is a versatile experimental technique for probing the electronic structure of materials. It provides a wealth of information on the sample's atomic-scale structure, but extracting this information from the experimental data can be challenging because there is no direct relation between the structure and the measured spectrum. Theoretical calculations can bridge this gap by explaining the structural origins of the spectral features. Reliable methods for modeling inelastic x-ray scattering require accurate electronic structure calculations. This work presents the development and implementation of new schemes for modeling the inelastic scattering of x-rays from non-periodic systems. The methods are based on density functional theory and are applicable for a wide variety of molecular materials. Applications are presented in this work for amorphous silicon monoxide and several gas phase systems. Valuable new information on their structure and properties could be extracted with the combination of experimental and computational methods.

Computational Methods for Studies of Quantum Dots and Boson Condensates

This thesis presents ab initio studies of two kinds of physical systems, quantum dots and bosons, using two program packages of which the bosonic one has mainly been developed by the author. The implemented models, \emph{i.e.}, configuration interaction (CI) and coupled cluster (CC) take the correlated motion of the particles into account, and provide a hierarchy of computational schemes, on top of which the exact solution, within the limit of the single-particle basis set, is obtained. The theory underlying the models is presented in some detail, in order to provide insight into the approximations made and the circumstances under which they hold. Some of the computational methods are also highlighted. In the final sections the results are summarized. The CI and CC calculations on multiexciton complexes in self-assembled semiconductor quantum dots are presented and compared, along with radiative and non-radiative transition rates. Full CI calculations on quantum rings and double quantum rings are also presented. In the latter case, experimental and theoretical results from the literature are re-examined and an alternative explanation for the reported photoluminescence spectra is found. The boson program is first applied on a fictitious model system consisting of bosonic electrons in a central Coulomb field for which CI at the singles and doubles level is found to account for almost all of the correlation energy. Finally, the boson program is employed to study Bose-Einstein condensates confined in different anisotropic trap potentials. The effects of the anisotropy on the relative correlation energy is examined, as well as the effect of varying the interaction potential.}