Meta-analysis in psychiatric genetics.
Meta-analysis in psychiatric genetics.
复制标题
DOI：
10.1007/s11920-005-0012-9
复制
发表时间：
2005-04-01
影响因子：
6.7
通讯作者：
Levinson, Douglas F
中科院分区：
文献类型：
--
作者：
Levinson, Douglas F
本文综述了遗传连锁和关联研究的meta分析方法的文献，并对精神疾病的具体meta分析结果进行了总结和评论。基因组扫描荟萃分析和多重扫描概率方法评估了研究间关联的证据。多重扫描概率分析表明，两个染色体区域（13q和22q）与精神分裂症和双相情感障碍有关，而基因组扫描荟萃分析在更大的样本中发现了至少10个精神分裂症连锁区域，但没有一个与双相情感障碍有关。汇集ORs的荟萃分析支持精神分裂症与DRD2中的Ser311Cys多态性和HTR2A中的T102C多态性相关，以及注意缺陷多动障碍与DRD4中的48bp重复序列相关。5-羟色胺转运基因（SLC6A4）中的5-HTTLPR多态性可能与双相情感障碍、自杀行为和神经质的风险有关，但与重度抑郁症终生风险的关系尚未得到证实。荟萃分析支持精神分裂症与通过定位克隆方法确定的候选基因可复制关联区域的联系。还有额外的支持区域，其中易感基因可能被确定。基于较小的数据集，关联荟萃分析在双相情感障碍方面的成功并不明显。荟萃分析可以指导研究区域的优先顺序，但证明关联需要对基因作用的假设进行生物学确认。阐明因果机制需要对候选基因的序列变异进行更全面的研究，需要更好的统计和荟萃分析方法来考虑所有变异，需要检验病原学假设的生物学策略。
The article reviews literature on methods for meta-analysis of genetic linkage and association studies, and summarizes and comments on specific meta-analysis findings for psychiatric disorders. The Genome Scan Meta-Analysis and Multiple Scan Probability methods assess the evidence for linkage across studies. Multiple Scan Probability analysis suggested linkage of two chromosomal regions (13q and 22q) to schizophrenia and bipolar disorder, whereas Genome Scan Meta-Analysis on a larger sample identified at least 10 schizophrenia linkage regions, but none for bipolar disorder. Meta-analyses of pooled ORs support association of schizophrenia to the Ser311Cys polymorphism in DRD2 and the T102C polymorphism in HTR2A, and of attention deficit hyperactivity disorder to the 48-bp repeat in DRD4. The 5-HTTLPR polymorphism in the serotonin transporter gene (SLC6A4) may contribute to the risk of bipolar disorder, suicidal behavior, and neuroticism, but association to the lifetime risk of major depression has not been shown. Meta-analyses support linkage of schizophrenia to regions where replicable associations to candidate genes have been identified through positional cloning methods. There are additional supported regions where susceptibility genes are likely to be identified. Linkage meta-analysis has had less clear success for bipolar disorder based on a smaller dataset. Meta-analysis can guide the prioritization of regions for study, but proof of association requires biological confirmation of hypotheses about gene actions. Elucidation of causal mechanisms will require more comprehensive study of sequence variation in candidate genes, better statistical and meta-analytic methods to take all variation into account, and biological strategies for testing etiologic hypotheses.