Do not imput onther's sin to yourself scheme of imputation in statistics GWAS 学习笔记—— imputation 的含义 (Truth of Imputation) by Baoyu 7 月份参加了 Fudan 主办的首期 GWAS 研修班，对国内外 GWAS 有了系统、深入的感知。学习过程中，需要精准地理解一些重要概念，才能避免一知半解。 GWAS 相关重要的名词有 Effect heterogeneity ， Pooling ， Replication\Joint\Pooled analysis ， Geographic variation ， Hierarchical Clustering ， imputation ， marginal effects ， Manhattan Forest ， GWAS consortium ( 如 Genetic Association Information Network, GAIN) 。这次班上不少老师把 Manhattan Forest 改称为 Pudong Forest ， Pudong 楼群的高度和密度都不亚于 Manhattan) ，呵呵。 GWAS 研究中， imputation 前承高通量测序，下启数据分析；对缺失数据的 imputation 是进行数据分析的前提，重要性不言而喻。 Impute 愿意是归罪 , 归咎 , 归因 , 非难 , 诋毁。 Merian-Webster 辞典的解释既是此义： Main Entry: impute Function: transitive verb Inflected Form:imputed ; imputing Etymology: Middle English inputen, from Latin imputare, from in- + putare to consider Date: 14th century 1: to l ay the responsibility or blame for often falsely or unjustly 2: to credit to a person or a cause : ATTRIBUTE *our vices as well as our virtues have been imputed to bodily derangement B. N. Cardozo synonyms see ASCRIBE. 统计遗传学中意为预测、插补，由已知的基因型预测未知的基因型并对缺失的数据进行补缺，如这句： This imputation method uses the dense genotype data available from the HapMap CEU samples and the linkage disequilibrium (LD) relationships of the SNPs to impute (predict) genotypes for a large number of SNPs that were not measured experimentally in our Finnish cases and controls. Statistical genetics 中 imputation 的三个主要作用： Allows testing of untyped variation ， Allows easy combination of data across genotyping platforms ， Provides complete data for analysis with multiple SNPs. 实现 imputation 的 常用软件 ： 1. IMPUTE Developed by Jonathan Marchini Nature Genetics, Advance online publication http://www.stats.ox.ac.uk/~marchini/#software 2. Mach 1.0, Markov Chain Haplotyping Developed by Goncalo Abecasis http://www.sph.umich.edu/csg/abecasis/MACH/ 附件 1 是 U Michigen 的小牛 Scott 对 imputation 等分析方法的一个介绍。 附件 2 是 Eric E Schadt (Rosetta Inpharmatics) 实验室最近在 BMC Genetis 上一篇题为 GWAS 中插补（ imputation ）准确度及对关联分析统计效力的影响，值得一读。 附件 1 Scott_Handling and analyzing data of GWAS 附件 2 09 Accuracy of genome-wide imputation of untyped m arkers and...
标签: 插补
