2020 Fiscal Year Final Research Report
Clinical Genetics Knowledge Sharing System for Congenital Rare Disease Deep Phenotyping Overcoming the "30% limit" of Diagnosis
| Project/Area Number |
18K07850
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 52050:Embryonic medicine and pediatrics-related
|
|---|
| Research Institution | Nagasaki University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2018-04-01 – 2021-03-31
|
| Keywords | 希少疾患 / 顔貌解析 / 機械学習 |
|---|
| Outline of Final Research Achievements |
Face2Gene (https://face2gene.com/), developed by FDNA, Inc. in the U.S., is a machine-learning tool to assist in the diagnosis of congenital morphology syndromes based on facial gestalts. We attempted to evaluate its performance using cases recruited in Japan (49 cases in 26 syndromes) (Mishima et al., J Hum Genet, 2019). As a result, the system successfully presented the top 10 syndromes in 85.7% of the learned syndromes, indicating that the system already has high performance. At the same time, we found that some of the learned syndromes failed to be presented. The reason for this result could be the insufficient number of training cases or the difference in the strength of the effect of racial background.
|
| Free Research Field |
バイオインフォマティクス
|
| Academic Significance and Societal Importance of the Research Achievements |
Face2Geneは、すでに日本人集団においても高い候補症候群提示能力を持つことが明らかになった。このことは、同システムが臨床遺伝専門医が、本邦における9,000症候群近くが知られる先天形態異常症候群からの診断を絞りこむ上で有用であることを示した。また、先天形態異常症候群の原因である単一遺伝子のバリエーションが顔貌表現型に与える影響は、多くの場合地理的集団による遺伝的バックグラウンドよりも十分に強いものであることが示唆された。
|
