Improving fine-mapping by modeling infinitesimal effects.
Improving fine-mapping by modeling infinitesimal effects.
复制标题
通过对无穷小的影响进行建模来改进精细映射。
DOI：
10.1038/s41588-023-01597-3
复制
发表时间：
2024
期刊：
影响因子：
30.8
通讯作者：
Finucane,HilaryK
中科院分区：
文献类型：
--
作者：
Cui,Ran;Elzur,RoyA;Kanai,Masahiro;Ulirsch,JacobC;Weissbrod,Omer;Daly,MarkJ;Neale,BenjaminM;Fan,Zhou;Finucane,HilaryK
精细定位旨在识别表型的因果遗传变异。贝叶斯精细映射算法（例如，SuSiE，FINEMAP，ABF和COJO-ABF）被广泛使用，但在真实的数据中评估后验概率校准仍然具有挑战性，其中可能存在模型错误指定，并且真正的因果变量未知。我们引入了复制失败率（RFR），一个通过下采样来评估精细映射一致性的指标。SuSiE、FINEMAP和COJO-ABF显示出较高的RFR，表明对其输出可能过度自信。仿真结果表明，非稀疏遗传结构可能导致误校准，而插补噪声，因果变量和质量控制过滤器的非均匀分布的影响很小。在这里，我们提出了SuSiE-inf和FINEMAP-inf，精细映射方法建模无穷小的影响，以及更少的更大的因果关系。我们的方法显示出改进的校准，RFR和功能丰富，竞争性召回和计算效率。值得注意的是，使用我们的方法的后验效应量大大增加了多基因风险评分的准确性超过SuSiE和FINEMAP。我们的工作改进了复杂性状的因果变异识别，这是人类遗传学的一个基本目标。
Fine-mapping aims to identify causal genetic variants for phenotypes. Bayesian fine-mapping algorithms (for example, SuSiE, FINEMAP, ABF and COJO-ABF) are widely used, but assessing posterior probability calibration remains challenging in real data, where model misspecification probably exists, and true causal variants are unknown. We introduce replication failure rate (RFR), a metric to assess fine-mapping consistency by downsampling. SuSiE, FINEMAP and COJO-ABF show high RFR, indicating potential overconfidence in their output. Simulations reveal that nonsparse genetic architecture can lead to miscalibration, while imputation noise, nonuniform distribution of causal variants and quality control filters have minimal impact. Here we present SuSiE-inf and FINEMAP-inf, fine-mapping methods modeling infinitesimal effects alongside fewer larger causal effects. Our methods show improved calibration, RFR and functional enrichment, competitive recall and computational efficiency. Notably, using our methods’ posterior effect sizes substantially increases polygenic risk score accuracy over SuSiE and FINEMAP. Our work improves causal variant identification for complex traits, a fundamental goal of human genetics.