No description available.

The United States is currently experiencing an epidemic of opioid use, use disorder, and overdose-related deaths. While studies have identified several loci that are associated with opioid use disorder (OUD) risk, the genetic basis for the ability to discontinue opioid use has not been investigated. Furthermore, very few studies have investigated the non-genetic factors that are predictive of opioid cessation or their predictive ability. In this thesis, I studied a novel phenotype–opioid cessation, defined as the time since last use of illicit opioids (< 6 months ago as not cease, >1 year ago as cease) among persons meeting lifetime DSM-5 criteria for opioid use disorder (OUD). In chapter two, I identified novel genetic variants and biological pathways that potentially regulate opioid cessation success through a genome wide study, as well as genetic overlap between opioid cessation and other substance cessation traits. In chapter three, I identified multiple non-genetic risk factors specific to each racial group that are predictive of opioid cessation from the same individuals analyzed in chapter two by applying several linear and non-linear machine learning techniques to a set of more than 3,000 variables assessed by a structured psychiatric interview. Factors identified from this atheoretical approach can be grouped into opioid use activities, other drug use, health conditions, and demographics, while the predictive accuracy as high as nearly 80% was achieved. The findings from this research generated more hypotheses for future studies to reference. In chapter four, I performed differential gene expression and network analysis on mice with different oxycodone (an opioid receptor agonist)-induced behaviors and compared the significantly associated genes and network modules with top-ranked genes identified in humans. The pathway cross-talks and gene homologs identified from both species illuminate the potential molecular mechanism of opioid behaviors. In summary, this thesis utilized statistical genetics, machine learning, and a computational biology framework to address factors that are associative with opioid cessation in humans, and cross-referenced the genetic findings in a mouse model. These findings serve as references for future studies and provide a framework for personalizing the treatment of OUD.

The influence of pollinators on the evolution of flower morphology has been extensively explored. Yet, the effect of other ecological factors, such as genetic drift, environmental filtering, and allometric constraints, gained less attention. In this study, we addressed the importance of those drivers in a predominantly selfing species. 400 worldwide Arabidopsis thaliana accessions were gathered and grown in semi-controlled climatic settings to explore the association between flower organ size, genotypes, and habitats. In our dataset, petal area was the most variable trait. Petal size was phenotypically and genetically correlated with other flowering structures, but no genetic allometry constraints were found to affect petal size evolution. The negative correlation of petal size with fitness and the traces of selective constraints in petal associated genes suggest that petal size is currently under selection in this species. We found paucity of genotypes harbouring large petals at low suitability regions, which points to the presence of environmental filtering. The novelty of this project relies on the pluralistic integration of factors studied and highlights the role of the climate on flower size evolution. Our results suggest that resource allocation is an important driver of flower size evolution in self-fertilising species but that its effect is largely determined by local environmental pressures.

In the last years, Genome-Wide Associations Studies (GWAS) found many variants associated with complex diseases. However, the biological and molecular links between these variants and phenotypes are still mostly unknown. Also, even if sample sizes are constantly increasing, the associated variants do not explain all the heritability estimated for many traits. Many hypotheses have been proposed to explain the problem: from variant-variant interactions, the effect of rare and ultra-rare coding variants and also technical biases related to sequencing or statistic on sexual chromosomes. In this thesis, we mainly explore the hypothesis of variant-variant interaction and, briefly, the rare coding variants hypothesis while also considering possible molecular effects like allele-specific expression and the effects of variants on protein interfaces. Some parts of the thesis are also devoted to explore the implementation of efficient computational tools to explore these effects and to perform scalable genotyping of germline single nucleotide polymorphisms (SNPs) in huge datasets. The main part of the thesis regards the development of a new resource to identify putative variant-variant interactions. In particular, we integrated ChIP-seq data from ENCODE, transcription factor binding motifs from several resources and genotype and transcript level data from GTeX and TCGA. This new dataset allows us to formalize new models, to make hypothesis and to find putative novel associations and interactions between (mainly non-coding) germline variants and phenotypes, like cancer-specific phenotypes. In particular, we focused on the characterization of breast cancer and Alzheimer’s Disease GWAS risk variants, looking for putative variants’ interactions. Recently, the study of rare variants has become feasible thanks to the biobanks that made available genotypes and clinical data of thousands of patients. We characterize and explore the possible effects of rare coding inherited polymorphisms on protein interfaces in the UKBioBank trying to understand if the change in structure of protein can be one of the causes of complex diseases. Another part of the thesis explores variants as causal molecular effect for allele-specific expression. In particular, we describe UTRs variants that can alter the post-transcriptional regulation in mRNA leading to a phenomenon where an allele is more expressed than the other. Finally, we show those variants can have prognostic significance in breast cancer. This thesis work introduces results and computational tools that can be useful to a broad community of researcher studying human polymorphisms effects.

Preferential fat accumulation in visceral vs. subcutaneous depots makes obese individuals more prone to metabolic complications. Body fat distribution (FD) is regulated by genetics. FD patterns vary across ethnic groups independent of obesity. Asians have more and Africans have less visceral fat compared with Europeans. Consequently, Asians tend to be more susceptible to type 2 diabetes even with lower BMIs when compared with Europeans. To date, genome-wide association studies (GWAS) have identified more than 460 loci related to FD traits. However, the majority of these data were generated in European populations. In this review, we aimed to summarize recent advances in FD genetics with a focus on comparisons between European and non-European populations (Asians and Africans). We therefore not only compared FD-related susceptibility loci identified in three ethnicities but also discussed whether known genetic variants might explain the FD pattern heterogeneity across different ancestries. Moreover, we describe several novel candidate genes potentially regulating FD, including NID2, HECTD4 and GNAS, identified in studies with Asian populations. It is of note that in agreement with current knowledge, most of the proposed FD candidate genes found in Asians belong to the group of developmental genes.

Holistic and growth stage-specific screening is needed for identifying tolerant genotypes and for formulating strategies to mitigate the negative effects of abiotic stresses on crops. The objectives of this study were to characterize the genetic variability of 100 rice lines for early-season vigor, growth and physiological plasticity, and drought and temperature tolerance. Five studies were conducted to accomplish these objectives. In study 1 and 2, 100 rice genotypes consisting of several cultivars and experimental breeding lines were characterized for early-season vigor using several shoot and root morphological, physiological, and yield related traits. In study 3, low- and high-temperature tolerance assessed on select rice cultivars/hybrids during early-season. In study 4, genotypic variability in response to drought stress tolerance using morpo-physiological traits including roots was assessed under pot-culture conditions in a mini-greenhouse conditions. In study 5, the 100 rice genotypes were used to identify and validate SNP markers, and genome-wide association study (GWAS) to generate genotypic and phenotypic data with the objective of identifying new genetic loci controlling drought stress traits. Significant variability was recorded among rice genotypes and treatments for many traits measured. Early-season cumulative vigor response indices (CVRI) developed by summing individual responses indices for each trait varied among the rice genotypes, 21.36 (RU1404196) to 36.17 (N-22). Based on means and standard deviation of the CVRI, rice genotypes were classified as low- (43) and moderately low- (33), high- (16), and very high-vigor (5) groups. Total low-temperature response index values ranged from 18.48 to 23.15 whereas total high-temperature responses index values ranged from 42.01 to 48.82. Antonio, CLXL 745, and Mermentau were identified as sensitive to cold- and heat, and XL 753 was highly cold and heat tolerant genotypes tested. A cumulative drought stress response index (CDSRI) values varied between 14.7 (CHENIERE) and 27.9 (RU1402174) among the genotypes tested. This preliminary analysis of GWA indicated that substantial phenotypic and genotypic diversity exists in the 100 rice genotypes, despite their narrow genetic pool. The stress tolerant and high vigor rice genotypes will be valuable for rice breeders for developing new genotypes best suited under growing environments prone to early-season drought and temperature.

<p>Complex disorders are caused by multiple genetic, environmental, and lifestyle factors, and their interactions. Most human diseases are complex, including many psychiatric, autoimmune, neurodegenerative, and cardiovascular disorders. Understanding their genetic background is an essential step toward developing effective preventive and therapeutic interventions for these disorders. In this dissertation, we present an overview of state-of-the-art methodology that is used to help elucidate the genetic basis of complex diseases and apply these methods to understand the genetic background of different complex disorders. First, we carried out a GWAS for myasthenia gravis (MG), a rare autoimmune disorder, and detected a novel risk locus, AGRN, which encodes a protein, involved in neuromuscular junction activation. Additionally, we observed significant genetic correlation between MG and ADs, and variants with pleiotropic effects. Second, we explored the genetic and phenotypic relationships among 11 different autoimmune disorders (ADs), using GWAS results o to calculate polygenic risk scores (PRS) and performing a PRS- phenome-wide association study (PheWAS) analysis with 3,281 phenotypes available in the UK Biobank. We observed associations of ADs PRS with phenotypes in multiple categories, including lifestyle, biomarkers, mental and physical health. We also explored the shared genetic components among the ADs, through genetic correlation and cross-disorder meta-analysis approaches, where we</p> <p>identified pleiotropic variants among the correlated ADs. Finally, we performed a meta-analysis GWAS of Tourette Syndrome (TS) followed by post-GWAS analyses including biological annotation of the results, and association tests of TS PRS with brain volumes. We detected a novel locus, NR2F1, associated with TS, supported by eQTL and Hi-C data. TS PRS was significantly associated with right and left thalamus volumes and right putamen volume. Overall, our work demonstrates the power of GWAS and related methods to help disentangle the genetic basis of complex disease and provides important insights into the genetic basis of the specific disorders that are the focus of our studies.</p>

No description available.

No description available.

Recent advances in biotechnologies have enabled multiplatform and large-scale quantitative measurements of biomedical events. The need to analyze the produced vast amount of imaging and genomic data stimulates various novel applications of statistical machine learning methods in many areas of biomedical research. The main objective is to assist biomedical investigators to better interpret, analyze, and understand the biomedical questions based on the acquired data. Given the computational challenges imposed by these high-dimensional and complex data, machine learning research finds its new opportunities and roles. In this dissertation thesis, we propose to develop, test and apply novel statistical machine learning methods to analyze the data mainly acquired by dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and single nucleotide polymorphism (SNP) microarrays. The research work focuses on: (1) tissue-specific compartmental analysis for dynamic contrast-enhanced MR imaging of complex tumors; (2) computational Analysis for detecting DNA SNP interactions in genome-wide association studies. DCE-MRI provides a noninvasive method for evaluating tumor vasculature patterns based on contrast accumulation and washout. Compartmental analysis is a widely used mathematical tool to model dynamic imaging data and can provide accurate pharmacokinetics parameter estimates. However partial volume effect (PVE) existing in imaging data would have profound effect on the accuracy of pharmacokinetics studies. We therefore propose a convex analysis of mixtures (CAM) algorithm to explicitly eliminate PVE by expressing the kinetics in each pixel as a nonnegative combination of underlying compartments and subsequently identifying pure volume pixels at the corners of the clustered pixel time series scatter plot. The algorithm is supported by a series of newly proved theorems and additional noise filtering and normalization preprocessing. We demonstrate the principle and feasibility of the CAM approach together with compartmental modeling on realistic synthetic data, and compare the accuracy of parameter estimates obtained using CAM or other relevant techniques. Experimental results show a significant improvement in the accuracy of kinetic parameter estimation. We then apply the algorithm to real DCE-MRI data of breast cancer and observe improved pharmacokinetics parameter estimation that separates tumor tissue into sub-regions with differential tracer kinetics on a pixel-by-pixel basis and reveals biologically plausible tumor tissue heterogeneity patterns. This method has combined the advantages of multivariate clustering, convex optimization and compartmental modeling approaches. Interactions among genetic loci are believed to play an important role in disease risk. Due to the huge dimension of SNP data (normally several millions in genome-wide association studies), the combinatorial search and statistical evaluation required to detect multi-locus interactions constitute a significantly challenging computational task. While many approaches have been proposed for detecting such interactions, their relative performance remains largely unclear, due to the fact that performance was evaluated on different data sources, using different performance measures, and under different experimental protocols. Given the importance of detecting gene-gene interactions, a thorough evaluation of the performance and limitations of available methods, a theoretical analysis of the interaction effect and the genetic factors it depends on, and the development of more efficient methods are warranted. Therefore, we perform a computational analysis for detect interactions among SNPs. The contributions are four-fold: (1) developed simulation tools for evaluating performance of any technique designed to detect interactions among genetic variants in case-control studies; (2) used these tools to compare performance of five popular SNP detection methods; and (3) derived analytic relationships between power and the genetic factors, which not only support the experimental results but also gives a quantitative linkage between interaction effect and these factors; (4) based on the novel insights gained by comparative and theoretical analysis, developed an efficient statistically-principled method, namely the hybrid correlation-based association (HCA) to detect interacting SNPs. The HCA algorithm is based on three correlation-based statistics, which are designed to measure the strength of multi-locus interaction with three different interaction types, covering a large portion of possible interactions. Moreover, to maximize the detection power (sensitivity) while suppressing false positive rate (or retaining moderate specificity), we also devised a strategy to hybridize these three statistics in a case-by-case way. A heuristic search strategy is also proposed to largely decrease the computational complexity, especially for high-order interaction detection. We have tested HCA in both simulation study and real disease study. HCA and the selected peer methods were compared on a large number of simulated datasets, each including multiple sets of interaction models. The assessment criteria included several power measures, family-wise type I error rate, and computational complexity. The experimental results of HCA on the simulation data indicate its promising performance in terms of a good balance between detection accuracy and computational complexity. By running on multiple real datasets, HCA also replicates plausible biomarkers reported in previous literatures. / Ph. D.