Li Danyang, Choque Olsson Nora, Becker Martin, Arora Abishek, Jiao Hong, Norgren Nina, Jonsson Ulf, Bölte Sven, Tammimies Kristiina
Center of Neurodevelopmental Disorders (KIND), Centre for Psychiatry Research, Department of Women's and Children's Health, Karolinska Institutet and Child and Adolescent Psychiatry, Stockholm Health Care Services, Stockholm County Council, Stockholm, Sweden.
Astrid Lindgren Children's Hospital, Karolinska University Hospital, Region Stockholm, Solna, Sweden.
Autism Res. 2022 Mar;15(3):434-446. doi: 10.1002/aur.2666. Epub 2021 Dec 30.
Exome sequencing has been proposed as the first-tier genetic testing in autism spectrum disorder (ASD). Here, we performed exome sequencing in autistic individuals with average to high intellectual abilities (N = 207) to identify molecular diagnoses and genetic modifiers of intervention outcomes of social skills group training (SSGT) or standard care. We prioritized variants of clinical significance (VCS), variants of uncertain significance (VUS) and generated a pilot scheme to calculate genetic scores of rare and common variants in ASD-related gene pathways. Mixed linear models were used to test the association between the carrier status of VCS/VUS or the genetic scores with intervention outcomes measured by the social responsiveness scale. Additionally, we combined behavioral and genetic features using a machine learning (ML) model to predict the individual response. We showed a rate of 4.4% and 11.3% of VCS and VUS in the cohort, respectively. Individuals with VCS or VUS had improved significantly less after standard care than non-carriers at post-intervention (β = 9.35; p = 0.036), while no such association was observed for SSGT (β = -2.50; p = 0.65). Higher rare variant genetic scores for synaptic transmission and regulation of transcription from RNA polymerase II were separately associated with less beneficial (β = 8.30, p = 0.0044) or more beneficial (β = -6.79, p = 0.014) effects after SSGT compared with standard care at follow-up, respectively. Our ML model showed the importance of rare variants for outcome prediction. Further studies are needed to understand genetic predisposition to intervention outcomes in ASD.
外显子组测序已被提议作为自闭症谱系障碍(ASD)的一级基因检测方法。在此,我们对智力水平中等至高的自闭症个体(N = 207)进行了外显子组测序,以确定分子诊断以及社交技能小组训练(SSGT)或标准护理干预结果的基因修饰因子。我们对具有临床意义的变异(VCS)、意义未明的变异(VUS)进行了优先级排序,并生成了一个初步方案来计算ASD相关基因途径中罕见和常见变异的基因分数。使用混合线性模型来测试VCS/VUS的携带状态或基因分数与通过社交反应量表测量的干预结果之间的关联。此外,我们使用机器学习(ML)模型结合行为和基因特征来预测个体反应。我们发现该队列中VCS和VUS的比例分别为4.4%和11.3%。在干预后,具有VCS或VUS的个体在接受标准护理后的改善明显低于非携带者(β = 9.35;p = 0.036),而在SSGT中未观察到这种关联(β = -2.50;p = 0.65)。与随访时的标准护理相比,突触传递和RNA聚合酶II转录调控的较高罕见变异基因分数分别与SSGT后较不利(β = 8.30,p = 0.0044)或更有利(β = -6.79,p = 0.014)的效果相关。我们的ML模型显示了罕见变异对结果预测的重要性。需要进一步研究以了解ASD干预结果的遗传易感性。