Fused multi-modal similarity network as prior in guiding brain imaging genetic association.
Fused multi-modal similarity network as prior in guiding brain imaging genetic association.
复制标题
DOI：
10.3389/fdata.2023.1151893
复制
发表时间：
2023
影响因子：
3.1
通讯作者：
Yan, Jingwen
中科院分区：
文献类型：
--
作者：
He, Bing;Xie, Linhui;Varathan, Pradeep;Nho, Kwangsik L.;Risacher, Shannon L. J.;Saykin, Andrew J.;Yan, Jingwen
脑成像遗传学旨在探索大脑结构和功能背后的遗传结构。最近的研究表明，结合先前的知识，如受试者诊断信息和大脑区域相关性，可以帮助识别更强的成像遗传关联。然而，有时这些信息可能是不完整的，甚至是不可用的。在这项研究中，我们探索了一种新的数据驱动的先验知识，通过融合多通道相似度网络来获取主题级相似度。它被纳入稀疏典型相关分析(SCCA)模型，该模型旨在识别一小部分大脑成像和遗传标记，这些标记解释了这两种模式支持的相似性矩阵。它分别应用于ADNI队列的淀粉样蛋白和tau成像数据。跨成像和遗传数据的融合相似度矩阵被发现比诊断信息更好或相似地改善了关联性能，因此当诊断信息不可用时(即，以健康对照为重点的研究)可能成为一种潜在的替代先验。我们的结果证实了所有类型的先验知识在提高关联识别方面的价值。此外，与诊断网络和共表达网络相比，表示多模式数据支持的主题关系的融合网络表现出一致的最佳或同等最佳的性能。
Brain imaging genetics aims to explore the genetic architecture underlying brain structure and functions. Recent studies showed that the incorporation of prior knowledge, such as subject diagnosis information and brain regional correlation, can help identify significantly stronger imaging genetic associations. However, sometimes such information may be incomplete or even unavailable. In this study, we explore a new data-driven prior knowledge that captures the subject-level similarity by fusing multi-modal similarity networks. It was incorporated into the sparse canonical correlation analysis (SCCA) model, which is aimed to identify a small set of brain imaging and genetic markers that explain the similarity matrix supported by both modalities. It was applied to amyloid and tau imaging data of the ADNI cohort, respectively. Fused similarity matrix across imaging and genetic data was found to improve the association performance better or similarly well as diagnosis information, and therefore would be a potential substitute prior when the diagnosis information is not available (i.e., studies focused on healthy controls). Our result confirmed the value of all types of prior knowledge in improving association identification. In addition, the fused network representing the subject relationship supported by multi-modal data showed consistently the best or equally best performance compared to the diagnosis network and the co-expression network.
登录
查看更多内容
影响因子：
16.2
作者：
Kim, Jungsu;Basak, Jacob M.;Holtzman, David M.
通讯作者：
Holtzman, David M.
影响因子：
14.9
作者：
Kuleshov MV;Jones MR;Rouillard AD;Fernandez NF;Duan Q;Wang Z;Koplev S;Jenkins SL;Jagodnik KM;Lachmann A;McDermott MG;Monteiro CD;Gundersen GW;Ma'ayan A
通讯作者：
Ma'ayan A
影响因子：
--
作者：
Shen L;Chepelev I;Liu J;Wang W
通讯作者：
Wang W
影响因子：
4.8
作者：
Insel, Philip S.;Mormino, Elizabeth C.;Donohue, Michael C.
通讯作者：
Donohue, Michael C.
影响因子：
3.7
作者：
Siddarth, Prabha;Burggren, Alison C.;Small, Gary W.
通讯作者：
Small, Gary W.