File Download

Conference Paper: Gene-Gene Interaction in disease association for systemic lupus erythematosus in Asian populations

TitleGene-Gene Interaction in disease association for systemic lupus erythematosus in Asian populations
Authors
Issue Date2012
PublisherThe American Society of Human Genetics.
Citation
The 62nd Annual Meeting of the American Society of Human Genetics (ASHG 2012), San Francisco, CA., 6-10 November 2012. How to Cite?
AbstractSystemic lupus erythematosus (SLE) is characterized as an autoimmune disorder with unclear etiology. Genome-wide association study (GWAS) has been proved to be a powerful approach for uncovering association between genetic variation and disease risk. Single locus analysis is widely used in most GWAS to date. However, all the confirmed susceptibility genes still only explain a small fraction of disease heritability. Gene-gene interaction may play a role in disease association, but has so far been missing in the bigger picture of connection between genetic factors and complex diseases. In this study, genome-wide gene-gene interaction was analyzed based on two GWAS data sets from two different regions in China, namely Hong Kong and Anhui. Three statistical methods installed in PLINK and MDR were used to calculate the genetic epistasis using logistic regression, machine learning and information gain theory. First, logistic regression was used for preliminary genome-wide selection of pairwise variant-variant interactions in the two independent GWAS datasets. A number of SNP pairs showed statistically significant P value on genetic interaction. These SNP pairs were kept for further validation if one or both of which are located in a region with annotated biological function. Selected SNP pairs were then analyzed by two MDR methods. By machine learning method, new variables were constructed to classify high risk and low risk genotype combinations. Further, information gain theory was used to detect whether the information between two SNPs were redundant. A number of variant pairs were found to have significant epistatic interaction in disease association in both Hong Kong and Anhui GWAS, such as interaction between SNPs in MTHFR and CXCR4. Interactions between established loci were also confirmed by our data, such as that between BANK1 and BLK. These findings are currently being replicated by a larger data set in independent cohorts.
DescriptionThe Meeting abstracts' web site is located at http://www.ashg.org/meetings/meetings_abstract_search.shtml
Persistent Identifierhttp://hdl.handle.net/10722/185073

 

DC FieldValueLanguage
dc.contributor.authorYang, Jen_US
dc.contributor.authorZhang, Yen_US
dc.contributor.authorZhang, Xen_US
dc.contributor.authorLau, YL-
dc.contributor.authorYang, W-
dc.date.accessioned2013-07-15T10:28:40Z-
dc.date.available2013-07-15T10:28:40Z-
dc.date.issued2012en_US
dc.identifier.citationThe 62nd Annual Meeting of the American Society of Human Genetics (ASHG 2012), San Francisco, CA., 6-10 November 2012.en_US
dc.identifier.urihttp://hdl.handle.net/10722/185073-
dc.descriptionThe Meeting abstracts' web site is located at http://www.ashg.org/meetings/meetings_abstract_search.shtml-
dc.description.abstractSystemic lupus erythematosus (SLE) is characterized as an autoimmune disorder with unclear etiology. Genome-wide association study (GWAS) has been proved to be a powerful approach for uncovering association between genetic variation and disease risk. Single locus analysis is widely used in most GWAS to date. However, all the confirmed susceptibility genes still only explain a small fraction of disease heritability. Gene-gene interaction may play a role in disease association, but has so far been missing in the bigger picture of connection between genetic factors and complex diseases. In this study, genome-wide gene-gene interaction was analyzed based on two GWAS data sets from two different regions in China, namely Hong Kong and Anhui. Three statistical methods installed in PLINK and MDR were used to calculate the genetic epistasis using logistic regression, machine learning and information gain theory. First, logistic regression was used for preliminary genome-wide selection of pairwise variant-variant interactions in the two independent GWAS datasets. A number of SNP pairs showed statistically significant P value on genetic interaction. These SNP pairs were kept for further validation if one or both of which are located in a region with annotated biological function. Selected SNP pairs were then analyzed by two MDR methods. By machine learning method, new variables were constructed to classify high risk and low risk genotype combinations. Further, information gain theory was used to detect whether the information between two SNPs were redundant. A number of variant pairs were found to have significant epistatic interaction in disease association in both Hong Kong and Anhui GWAS, such as interaction between SNPs in MTHFR and CXCR4. Interactions between established loci were also confirmed by our data, such as that between BANK1 and BLK. These findings are currently being replicated by a larger data set in independent cohorts.-
dc.languageengen_US
dc.publisherThe American Society of Human Genetics.-
dc.relation.ispartofAnnual Meeting of the American Society of Human Genetics, ASHG 2012en_US
dc.titleGene-Gene Interaction in disease association for systemic lupus erythematosus in Asian populationsen_US
dc.typeConference_Paperen_US
dc.identifier.emailYang, J: jingy09@hku.hken_US
dc.identifier.emailLau, YL: lauylung@hku.hken_US
dc.identifier.emailYang, W: yangwl@hkucc.hku.hken_US
dc.identifier.authorityLau, YL=rp00361en_US
dc.identifier.authorityYang, W=rp00524en_US
dc.description.naturelink_to_OA_fulltext-
dc.identifier.hkuros215376en_US
dc.publisher.placeUnited States-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats