Leveraging network analytics to infer patient syndrome and identify causal genes in rare disease cases

BMC Genomics

Research

Susan Tang¹  Daniel Rene Richards¹  Sohela Shah¹  Robert Anthony Rebres¹  Andreas Krämer¹ 

[1] QIAGEN Bioinformatics, 1001 Marshall Street, Suite 200, 94063, Redwood City, CA, USA;

关键词: NGS;    Whole-genome sequencing;    Exome sequencing;    Rare disease diagnosis;    Variant selection;    Genetic disorders;    Diagnostic odyssey;   

DOI  :  10.1186/s12864-017-3910-4

来源: Springer

【 摘 要 】

BackgroundNext-generation sequencing is widely used to identify disease-causing variants in patients with rare genetic disorders. Identifying those variants from whole-genome or exome data can be both scientifically challenging and time consuming. A significant amount of time is spent on variant annotation, and interpretation. Fully or partly automated solutions are therefore needed to streamline and scale this process.ResultsWe describe Phenotype Driven Ranking (PDR), an algorithm integrated into Ingenuity Variant Analysis, that uses observed patient phenotypes to prioritize diseases and genes in order to expedite causal-variant discovery. Our method is based on a network of phenotype-disease-gene relationships derived from the QIAGEN Knowledge Base, which allows for efficient computational association of phenotypes to implicated diseases, and also enables scoring and ranking.ConclusionsWe have demonstrated the utility and performance of PDR by applying it to a number of clinical rare-disease cases, where the true causal gene was known beforehand. It is also shown that PDR compares favorably to a representative alternative tool.

【 授权许可】

CC BY   
© The Author(s). 2017

【 图 表 】

12864_2017_3910_Article_IEq2.gif

12864_2017_3910_Article_IEq1.gif