Combining Artificial Intelligence and RNA-Seq to elucidate new etiologies of genetic neurological disorders in childhood.

2023 Fiscal Year Final Research Report

• Project/Area Number: 20K08236
• 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: Showa University
• Principal Investigator: Kato Mitsuhiro 昭和大学, 医学部, 教授 (10292434)
• Co-Investigator(Kenkyū-buntansha): 宮 冬樹 慶應義塾大学, 医学部(信濃町), 准教授 (50415311) ; 中村 和幸 山形大学, 医学部, 助教 (20436215)
• Project Period (FY): 2020-04-01 – 2024-03-31
• Keywords: エクソーム / スプライス異常 / SpliceAI / 人工知能 / 小児神経 / 遺伝子解析 / ゲノム

Outline of Final Research Achievements

Rare and intractable neurological diseases with childhood onset are highly heritable. However, even with whole genome sequencing, the causative gene identification rate is around 60%. New analysis methods need to be developed. Using SpliceAI, a splice site prediction algorithm based on artificial intelligence (AI) technology, we detected variants that had not been detected by conventional splice site prediction algorithms. An expression study of the variants using LCL showed aberrant transcripts suggesting splice abnormalities. Exome data of 488 samples from the probands and their families were analyzed using SpliceAI. Seven pathogenic splicing variants were identified. We confirmed the usefulness of AI in detecting new genetic etiologies.

Free Research Field

小児神経学

Academic Significance and Societal Importance of the Research Achievements

子どもの神経、特にてんかんや発達に影響する脳の稀な遺伝性疾患の原因として、それまで解析が進んでいなかったスプライス異常に着目した。人工知能(AI)を用いたスプライス部位の予測アルゴリズムSpliceAIを用いてそれまで同定されなかったスプライス異常をきたす数多くの病因変異を同定した。AIを用いることで、疾患原因としての遺伝子変異同定率を改善し、原因不明例を減少させることができた。