Raphael Recanati Genetic Institute, Rabin Medical Center-Beilinson Hospital, Zeev Jabotinsky 39, Petach Tikva 4941492, Israel.
Clalit Research Institute, Clalit Health Services, Tuval 40, Ramat Gan 5252247, Israel.
Database (Oxford). 2024 Jul 23;2024. doi: 10.1093/database/baae065.
Targeted gene panel sequencing is used to limit the search for causative genetic variants solely to genes with an established association with the phenotype. The design of gene panels is challenging due to the lack of consensus regarding phenotypic associations for some genes, which results in high variation in gene composition for the same panel offered by different laboratories. We developed PANGEN, a platform that provides a centralized resource for gene panel information, with the ability to compare and generate new intelligent diagnostic panels. Gene-phenotype associations were collected from 12 public and commercial sources (Blueprint, Cegat, Centogene, ClinGen, Fulgent, GeneDx, Health in Code, Human Phenotype Ontology, Invitae, PanelApp, Prevention genetics, and Pronto diagnostics). Gene-phenotype associations are categorized into tiers according to categories derived from the original source panel. Pairwise panel similarity was calculated by dividing the number of common genes by the total number of genes in both panels. Regions with extreme guanine-cytosine (GC) content were collected from the Genome in a Bottle stratifications dataset, and putative genomic duplications were retrieved from the University of Santa Cruz database. Overall, 1533 panels, 9759 phenotypes, and 6979 genes were collected. The platform provides an interface to (i) explore and compare collected panels, (ii) find similar panels, (iii) identify genes with high GC content or duplication levels, (iv) generate gene panels by combining panels from various sources, and (v) stratify a generated panel into genes with a strong phenotype association ('core') and those with a weaker association ('extended'). The presented platform represents a unique resource for gene panel exploration and comparison that facilitates the generation of tailored diagnostic panels through a public online web server. Database URL: https://c-gc.shinyapps.io/PANGEN/.
靶向基因panel 测序用于将致病基因变异的搜索仅限于与表型具有明确关联的基因。由于某些基因的表型关联缺乏共识，基因panel 的设计具有挑战性，这导致不同实验室提供的相同panel 的基因组成存在很大差异。我们开发了 PANGEN，这是一个提供基因panel 信息集中资源的平台，具有比较和生成新的智能诊断panel 的能力。从 12 个公共和商业来源（Blueprint、Cegat、Centogene、ClinGen、Fulgent、GeneDx、Health in Code、Human Phenotype Ontology、Invitae、PanelApp、Prevention genetics 和 Pronto diagnostics）收集基因-表型关联。根据源自原始panel 的类别，将基因-表型关联分为不同的层次。通过将两个panel 中共有的基因数量除以两个panel 中基因总数来计算panel 间的相似性。从基因组瓶 Stratifications 数据集收集具有极端鸟嘌呤-胞嘧啶 (GC) 含量的区域，并从圣克鲁兹大学数据库检索推定的基因组重复。总共收集了 1533 个panel、9759 个表型和 6979 个基因。该平台提供了一个接口，用于（i）探索和比较收集的panel，（ii）查找相似的panel，（iii）识别具有高 GC 含量或重复水平的基因，（iv）通过组合来自不同来源的 panel 生成基因 panel，（v）将生成的 panel 分层为具有强表型关联的基因（'核心'）和关联较弱的基因（'扩展'）。该平台代表了一个独特的基因panel 探索和比较资源，通过公共在线网络服务器方便地生成定制化的诊断panel。数据库网址：https://c-gc.shinyapps.io/PANGEN/。
