Informatics and molecular studies of SNPs, in relation to metabolic and cardiovascular phenotypes

Abstract

Due to the nature of complex diseases, many claimed associations between genes and phenotypes in the literature are uncertain. A set of computer programs Per1 scripts were written to go through Medline abstracts and retrieve information about gene phenotype associations. The computer program identifies gene-phenotype associations with specificity of 83% precision of 68% and a balance F measure of 75%. The summary of previous studies makes it possible to select candidate genes and found prior hypotheses.

The genes LTA, TNF (located on Chr 6) and LGALS2 (located Chr 22) were tested for their association with metabolic syndrome traits and myocardial infarction. Four SNPs in LTA (+80/81, +252, T26N) two in TNF (-238, -308) and one tagSNP in LGALS2 (rs7291467) were genotyped in a cohort of 3500 British women aged 60 to 79 (BWHHS). All SNPs were analysed using linear regression. The LTA and TNF SNPs were also analysed together as haplotypes. Rs7291467 was found to be associated with insulin-glucose profile, the mean difference in fasting insulin per minor allele was -4% (p=0.01 for trend by allele) and the mean difference in fasting glucose per minor allele was -1% (p=0.02 for trend by allele). An LTA haplotype *2 (31%) (TGT-Thr) was associated with a decrease of 0.4 BMI units and 0.064 mM triglycerides compared to *1 (37%) (GGC-Asn) (p=0.01). Analysed by itself a haplotype harbouring TNF -308 A is associated with lower BMI but analysed together with LTA this association is explained by LTA +252 C. No associations were found with coronary heart disease.

More information

Identifiers

Catalogue record

Export record

ASCII CitationAtomBibTeXData Cite XMLDublin CoreDublin CoreEP3 XMLEndNoteHTML CitationHTML CitationHTML ListJSONMETSMODSMPEG-21 DIDLOpenURL ContextObjectOpenURL ContextObject in SpanRDF+N-TriplesRDF+N3RDF+XMLRIOXX2 XMLReferReference ManagerSimple Metadata