Overlapping but asymmetrical relationships between schizophrenia and autism revealed by brain connectivity-猫眼课题宝

DOI：

10.1101/403212

发表时间：

2018-09

期刊：

bioRxiv

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Y. Yoshihara;G. Lisi;N. Yahata;J. Fujino;Yukiko Matsumoto;J. Miyata;Genichi Sugihara;S. Urayama;M. Kubota;Masahiro Yamashita;R. Hashimoto;N. Ichikawa;W. Cahn;N. V. van Haren;S. Mori;Y. Okamoto;K. Kasai;N. Kato;H. Imamizu;R. Kahn;A. Sawa;M. Kawato;T. Murai;J. Morimoto;Hidehiko Takahashi

虽然精神分裂症谱系障碍（SSD）与自闭症谱系障碍（ASD）之间的关系一直存在争议，但尚未完全阐明。为了解决这个问题，我们利用了双重（ASD和SSD）分类器，根据静息状态脑功能连接将患者与对照组区分开来。使用复杂的机器学习算法自动选择SSD特定功能连接的SSD分类器应用于日本数据集，包括慢性SSD成年患者。我们在独立验证队列中证明了SSD分类的良好性能。在美国和欧洲的慢性队列中，以及一个包括首发精神分裂症的美国队列中，测试了这种普遍性。两组日本成年ASD和重度抑郁症患者以及一组欧洲注意力缺陷多动障碍患者测试了这种特异性。分类器功能连接的加权线性总和构成了代表神经对疾病的倾向性的生物学维度。我们之前开发的鲁棒ASD分类器构成了ASD维度。比较SSD、ASD和健康对照的SSD和ASD生物学维度的分布。SSD和ASD人群在两个生物学维度上表现出重叠但不对称的模式。也就是说，SSD人群在ASD维度上表现出更高的倾向性，反之则不然。此外，这两个维度在ASD人群中相关，而在SSD人群中不相关。基于静息状态功能连通性的两个生物学维度使我们能够量化和可视化SSD和ASD之间的关系。

Although the relationship between schizophrenia spectrum disorder (SSD) and autism spectrum disorder (ASD) has long been debated, it has not yet been fully elucidated. To address this issue, we took advantage of dual (ASD and SSD) classifiers that discriminate patients from their controls based on resting state brain functional connectivity. An SSD classifier using sophisticated machine-learning algorithms that automatically selected SSD- specific functional connections was applied to Japanese datasets including adult patients with SSD in a chronic stage. We demonstrated good performance of the SSD classification for independent validation cohorts. The generalizability was tested by USA and European cohorts in a chronic stage, and one USA cohort including first episode schizophrenia. The specificity was tested by two adult Japanese cohorts of ASD and major depressive disorder, and one European cohort of attention-deficit hyperactivity disorder. The weighted linear summation of the classifier’s functional connections constituted the biological dimensions representing neural liability to the disorders. Our previously developed robust ASD classifier constituted the ASD dimension. Distributions of individuals with SSD, ASD and healthy controls were examined on the SSD and ASD biological dimensions. The SSD and ASD populations exhibited overlapping but asymmetrical patterns on the two biological dimensions. That is, the SSD population showed increased liability on the ASD dimension, but not vice versa. Furthermore, the two dimensions were correlated within the ASD population but not the SSD population. Using the two biological dimensions based on resting-state functional connectivity enabled us to quantify and visualize the relationships between SSD and ASD.

Overlapping but asymmetrical relationships between schizophrenia and autism revealed by brain connectivity

文献摘要

相似文献