Google DeepMind推出AlphaGenome Atlas数据库，利用AI模型预测人类基因组中9亿个单核苷酸变异的调控影响1。该数据库规模达1个PB（petabyte），引入新的AlphaGenome Variant Impact（AVI）评分系统，融合了编码和非编码区域的预测能力，帮助科研人员快速筛选和优先处理具有研究前景的遗传变异1。
该工具已在多个实际应用中取得成效。Broad Institute研究团队利用AVI评分在DNM1基因中发现了关键变异，成功解决了一例罕见病诊断案例1。在复杂性状研究方面，研究员Dr. Gareth Hawkes在超过54,000名英国生物库参与者的数据中发现了22%更多的非编码遗传关联，识别出19个与BMI相关的遗传区域1。
AlphaGenome Atlas通过网页门户免费向全球研究人员开放，用户无需具备编码技能即可访问和使用1。
Google DeepMind has unveiled AlphaGenome Atlas, a comprehensive database leveraging artificial intelligence to forecast the functional impact of genetic variations across the human genome 1. The resource predicts the regulatory effects of approximately 9 billion single-letter genetic changes and comprises a dataset of 1 petabyte in scale 1.
The platform introduces the AlphaGenome Variant Impact (AVI) score, a novel scoring system that integrates predictions for both coding and non-coding genomic regions 1. AlphaGenome Atlas is freely accessible through a web portal and requires no coding expertise to use 1. Early applications have demonstrated its utility: researchers at the Broad Institute employed the AVI score to identify critical variants in the DNM1 gene, successfully resolving cases of rare genetic disorders 1. In parallel work, Dr. Gareth Hawkes analyzed data from over 54,000 UK Biobank participants and discovered 22 percent more non-coding genetic associations than previously identified, pinpointing 19 genetic regions linked to body mass index 1.
评论
还没有评论，欢迎留下第一条。