Eigen scores for human genome assembly GRCh38 Part 1 (Chr12 - Chr22)

Eigen is a spectral approach to the functional annotation of genetic variants in coding and noncoding regions. Eigen makes use of a variety of functional annotations in both coding and noncoding regions (such as protein function scores, evolutionary conservation scores, and epigenetic annotations from ENCODE and Roadmap Epigenomics projects), and combines them into one single measure of functional importance. Eigen is an unsupervised approach, and, unlike many existing methods, is not based on any labelled training data. Eigen produces estimates of predictive accuracy for each functional annotation score, and subsequently uses these estimates of accuracy to derive the aggregate functional score for variants of interest as a weighted linear combination of individual annotations.Eigen是一种用于编码区与非编码区遗传变异功能注释的谱方法。该方法整合了编码区与非编码区的多类功能注释信息，例如蛋白质功能评分、进化保守性评分，以及来自ENCODE和Roadmap表观基因组计划的表观基因组注释，并将其整合为单一的功能重要性度量指标。Eigen属于无监督方法，与多数现有方法不同，它无需依赖任何带标签的训练数据。该方法会对每项功能注释评分的预测准确率进行估算，随后基于这些准确率估算值，以各单项注释的加权线性组合形式，为目标变异推导得到综合功能评分。