Genome-wide association studies (GWASs) have received an increasing attention to understand genotype-phenotype relationships. The Bayesian network has been proposed as a powerful tool for modeling single-nucleotide polymorphism (SNP)-trait associations due to its advantage in addressing the high computational complex and high dimensional problems. Most current works learn the interactions among genotypes and phenotypes from the raw genotype data. However, due to the privacy issue, genotype information is sensitive and should be handled by complying with specific restrictions. In this work, we aim to build Bayesian networks from publicly released GWAS statistics to explicitly reveal the conditional dependency between SNPs and traits.
First, we focus on building a Bayesian network for modeling the SNP-categorical trait relationships. We construct a three-layered Bayesian network explicitly revealing the conditional dependency between SNPs and categorical traits from GWAS statistics. We then formulate inference problems based on the dependency relationship captured in the Bayesian network. Empirical evaluations show the effectiveness of our methods.
Second, we focus on modeling the SNP-quantitative trait relationships. Existing methods in the literature can only deal with categorical traits. We address this limitation by leveraging the Conditional Linear Gaussian (CLG) Bayesian network, which can handle a mixture of discrete and continuous variables. A two-layered CLG Bayesian network is built where the SNPs are represented as discrete variables in one layer and quantitative traits are represented as continuous variables in another layer. Efficient inference methods are then derived based on the constructed network. The experimental results demonstrate the effectiveness of our methods.
Finally, we present STIP, a web-based SNP-trait inference platform capable of a variety of inference tasks, such as trait inference given SNP genotypes and genotype inference given traits. The current version of STIP provides three services which are SNP-trait inference, Top-k trait prediction and GWAS catalog exploration.
","abstract_html":"<p>Genome-wide association studies (GWASs) have received an increasing attention to understand genotype-phenotype relationships. The Bayesian network has been proposed as a powerful tool for modeling single-nucleotide polymorphism (SNP)-trait associations due to its advantage in addressing the high computational complex and high dimensional problems. Most current works learn the interactions among genotypes and phenotypes from the raw genotype data. However, due to the privacy issue, genotype information is sensitive and should be handled by complying with specific restrictions. In this work, we aim to build Bayesian networks from publicly released GWAS statistics to explicitly reveal the conditional dependency between SNPs and traits. </p> <p>First, we focus on building a Bayesian network for modeling the SNP-categorical trait relationships. We construct a three-layered Bayesian network explicitly revealing the conditional dependency between SNPs and categorical traits from GWAS statistics. We then formulate inference problems based on the dependency relationship captured in the Bayesian network. Empirical evaluations show the effectiveness of our methods. </p> <p>Second, we focus on modeling the SNP-quantitative trait relationships. Existing methods in the literature can only deal with categorical traits. We address this limitation by leveraging the Conditional Linear Gaussian (CLG) Bayesian network, which can handle a mixture of discrete and continuous variables. A two-layered CLG Bayesian network is built where the SNPs are represented as discrete variables in one layer and quantitative traits are represented as continuous variables in another layer. Efficient inference methods are then derived based on the constructed network. The experimental results demonstrate the effectiveness of our methods. </p> <p>Finally, we present STIP, a web-based SNP-trait inference platform capable of a variety of inference tasks, such as trait inference given SNP genotypes and genotype inference given traits. The current version of STIP provides three services which are SNP-trait inference, Top-k trait prediction and GWAS catalog exploration. </p>","abstract_has_math":false,"creators":["Pan, Qiuping"],"institution":null,"degree_name":"Master of Science in Computer Science (MS)","degree_level":"Thesis","degree_discipline":null,"degree_department":null,"school":null,"contributors":["Li, Wing Ning","Li, Qinghua"],"advisors":["Wu, Xintao"],"committee_chairs":[],"committee_members":[],"year":2018,"date_issued":"2018-05-01T07:00:00Z","date_published":"2018-05-01T07:00:00Z","updated_at":"2026-07-24T00:58:14Z","subjects":["Bayesian Network","GWAS","GWAS Catalog","STIP","Bioinformatics","Genomics","OS and Networks"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://scholarworks.uark.edu/etd/2709","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Li, Wing Ning","Li, Qinghua"]},{"key":"dc:contributor.advisor","label":"Advisor","values":["Wu, Xintao"]},{"key":"dc:creator","label":"Author","values":["Pan, Qiuping"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2018"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2024-02-06T08:00:00Z"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science in Computer Science (MS)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Bayesian Network","GWAS","GWAS Catalog","STIP","Bioinformatics","Genomics","OS and Networks"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://scholarworks.uark.edu/etd/2709"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Genome-wide association studies (GWASs) have received an increasing attention to understand genotype-phenotype relationships. The Bayesian network has been proposed as a powerful tool for modeling single-nucleotide polymorphism (SNP)-trait associations due to its advantage in addressing the high computational complex and high dimensional problems. Most current works learn the interactions among genotypes and phenotypes from the raw genotype data. However, due to the privacy issue, genotype information is sensitive and should be handled by complying with specific restrictions. In this work, we aim to build Bayesian networks from publicly released GWAS statistics to explicitly reveal the conditional dependency between SNPs and traits.
First, we focus on building a Bayesian network for modeling the SNP-categorical trait relationships. We construct a three-layered Bayesian network explicitly revealing the conditional dependency between SNPs and categorical traits from GWAS statistics. We then formulate inference problems based on the dependency relationship captured in the Bayesian network. Empirical evaluations show the effectiveness of our methods.
Second, we focus on modeling the SNP-quantitative trait relationships. Existing methods in the literature can only deal with categorical traits. We address this limitation by leveraging the Conditional Linear Gaussian (CLG) Bayesian network, which can handle a mixture of discrete and continuous variables. A two-layered CLG Bayesian network is built where the SNPs are represented as discrete variables in one layer and quantitative traits are represented as continuous variables in another layer. Efficient inference methods are then derived based on the constructed network. The experimental results demonstrate the effectiveness of our methods.
Finally, we present STIP, a web-based SNP-trait inference platform capable of a variety of inference tasks, such as trait inference given SNP genotypes and genotype inference given traits. The current version of STIP provides three services which are SNP-trait inference, Top-k trait prediction and GWAS catalog exploration.
"]},{"key":"dc:title","label":"Title","values":["Bayesian Network Modeling and Inference of GWAS Catalog"]}]}],"canonical_facts":{"dc:contributor":["Li, Wing Ning","Li, Qinghua"],"dc:contributor.advisor":["Wu, Xintao"],"dc:creator":["Pan, Qiuping"],"dc:date":["2018"],"dc:date.available":["2024-02-06T08:00:00Z"],"dc:description.abstract":["Genome-wide association studies (GWASs) have received an increasing attention to understand genotype-phenotype relationships. The Bayesian network has been proposed as a powerful tool for modeling single-nucleotide polymorphism (SNP)-trait associations due to its advantage in addressing the high computational complex and high dimensional problems. Most current works learn the interactions among genotypes and phenotypes from the raw genotype data. However, due to the privacy issue, genotype information is sensitive and should be handled by complying with specific restrictions. In this work, we aim to build Bayesian networks from publicly released GWAS statistics to explicitly reveal the conditional dependency between SNPs and traits.
First, we focus on building a Bayesian network for modeling the SNP-categorical trait relationships. We construct a three-layered Bayesian network explicitly revealing the conditional dependency between SNPs and categorical traits from GWAS statistics. We then formulate inference problems based on the dependency relationship captured in the Bayesian network. Empirical evaluations show the effectiveness of our methods.
Second, we focus on modeling the SNP-quantitative trait relationships. Existing methods in the literature can only deal with categorical traits. We address this limitation by leveraging the Conditional Linear Gaussian (CLG) Bayesian network, which can handle a mixture of discrete and continuous variables. A two-layered CLG Bayesian network is built where the SNPs are represented as discrete variables in one layer and quantitative traits are represented as continuous variables in another layer. Efficient inference methods are then derived based on the constructed network. The experimental results demonstrate the effectiveness of our methods.
Finally, we present STIP, a web-based SNP-trait inference platform capable of a variety of inference tasks, such as trait inference given SNP genotypes and genotype inference given traits. The current version of STIP provides three services which are SNP-trait inference, Top-k trait prediction and GWAS catalog exploration.
"],"dc:identifier":["https://scholarworks.uark.edu/etd/2709"],"dc:subject":["Bayesian Network","GWAS","GWAS Catalog","STIP","Bioinformatics","Genomics","OS and Networks"],"dc:title":["Bayesian Network Modeling and Inference of GWAS Catalog"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["Master of Science in Computer Science (MS)"]},"updated_at":"2026-07-24T00:58:14Z"}