TY - JOUR
T1 - Meta-analysis of four new genome scans for lipid parameters and analysis of positional candidates in positive linkage regions
AU - Heijmans, B.T.
AU - Beekman, M.
AU - Putter, H.
AU - Lakenberg, N.
AU - van der Wijk, H.J.
AU - Whitfield, J.B.
AU - Posthuma, D.
AU - Pedersen, N.L.
AU - Martin, N.G.
AU - Boomsma, D.I.
AU - Slagboom, P.E.
PY - 2005
Y1 - 2005
N2 - Lipid levels in plasma strongly influence the risk for coronary heart disease. To localise and subsequently identify genes affecting lipid levels, we performed four genome-wide linkage scans followed by combined linkage/association analysis. Genome-scans were performed in 701 dizygotic twin pairs from four samples with data on plasma levels of HDL- and LDL-cholesterol and their major protein constituents, apolipoprotein AI (ApoAI) and Apolipoprotein B (ApoB). To maximise power, the genome scans were analysed simultaneously using a well-established meta-analysis method that was newly applied to linkage analysis. Overall LOD scores were estimated using the means of the sample-specific quantitative trait locus (QTL) effects inversely weighted by the standard errors obtained using an inverse regression method. Possible heterogeneity was accounted for with a random effects model. Suggestive linkage for HDL-C was observed on 8p23.1 and 12q21.2 and for ApoAI on 1q21.3. For LDL-C and ApoB, linkage regions frequently coincided (2p24.1, 2q32.1, 19p13.2 and 19q13.31). Six of the putative QTLs replicated previous findings. After fine mapping, three maximum LOD scores mapped within 1cM of major candidate genes, namely APOB (LOD =2.1), LDLR (LOD =1.9) and APOE (LOD =1.7). APOB haplotypes explained 27% of the QTL effect observed for LDL-C on 2p24.1 and reduced the LOD-score by 0.82. Accounting for the effect of the LDLR and APOE haplotypes did not change the LOD score close to the LDLR gene but abolished the linkage signal at the APOE gene. In conclusion, application of a new meta-analysis approach maximised the power to detect QTLs for lipid levels and improved the precision of their location estimate. © 2005 Nature Publishing Group. All rights reserved.
AB - Lipid levels in plasma strongly influence the risk for coronary heart disease. To localise and subsequently identify genes affecting lipid levels, we performed four genome-wide linkage scans followed by combined linkage/association analysis. Genome-scans were performed in 701 dizygotic twin pairs from four samples with data on plasma levels of HDL- and LDL-cholesterol and their major protein constituents, apolipoprotein AI (ApoAI) and Apolipoprotein B (ApoB). To maximise power, the genome scans were analysed simultaneously using a well-established meta-analysis method that was newly applied to linkage analysis. Overall LOD scores were estimated using the means of the sample-specific quantitative trait locus (QTL) effects inversely weighted by the standard errors obtained using an inverse regression method. Possible heterogeneity was accounted for with a random effects model. Suggestive linkage for HDL-C was observed on 8p23.1 and 12q21.2 and for ApoAI on 1q21.3. For LDL-C and ApoB, linkage regions frequently coincided (2p24.1, 2q32.1, 19p13.2 and 19q13.31). Six of the putative QTLs replicated previous findings. After fine mapping, three maximum LOD scores mapped within 1cM of major candidate genes, namely APOB (LOD =2.1), LDLR (LOD =1.9) and APOE (LOD =1.7). APOB haplotypes explained 27% of the QTL effect observed for LDL-C on 2p24.1 and reduced the LOD-score by 0.82. Accounting for the effect of the LDLR and APOE haplotypes did not change the LOD score close to the LDLR gene but abolished the linkage signal at the APOE gene. In conclusion, application of a new meta-analysis approach maximised the power to detect QTLs for lipid levels and improved the precision of their location estimate. © 2005 Nature Publishing Group. All rights reserved.
U2 - 10.1038/sj.ejhg.5201466
DO - 10.1038/sj.ejhg.5201466
M3 - Article
SN - 1018-4813
VL - 13
SP - 1143
EP - 1153
JO - European Journal of Human Genetics
JF - European Journal of Human Genetics
IS - 10
ER -