| 魏凤仙,武杰,杨叶,谢忠翔.基于组稀疏典型相关分析方法的影像遗传学方法在精神分裂症中的应用[J].中国医学影像技术,2019,35(2):277~281 |
| 基于组稀疏典型相关分析方法的影像遗传学方法在精神分裂症中的应用 |
| Application of imaging genetics method based on group sparse canonical correlation analysis in schizophrenia |
| 投稿时间：2018-07-13 修订日期：2018-10-22 |
| DOI：10.13929/j.1003-3289.201807106 |
| 中文关键词: 精神分裂症 稀疏表示 典型相关分析 单核苷酸多态性 |
| 英文关键词:schizophrenia sparse representation canonical correlation analysis single nucleotide polymorphism |
| 基金项目: |
|
| 摘要点击次数: 1913 |
| 全文下载次数: 1159 |
| 中文摘要: |
| 目的 采用影像遗传学研究方法探索精神分裂症患者的影像学数据与遗传学数据间的相关性。方法 提出一种组稀疏典型相关分析方法，在稀疏典型相关分析模型的基础上增加组稀疏惩罚项λ1‖u‖G和λ2‖v‖G进行变量组选择；对于选择的特征组，再利用组内惩罚项τ1‖u‖1和τ2‖v‖1进行组内的变量选择。采用基于组稀疏典型相关分析方法的影像遗传学方法分析精神分裂症患者脑区与相关基因位点的相关性，并验证其稳定性和筛选生物标记物的能力。结果 采用组稀疏典型相关分析方法获得了多组精神分裂症相关脑区和基因，其中左侧脑岛与基因AKT1的相关性最大，相关系数为0.653 8；右侧直回与基因DAOA和MAGI2的相关系数均大于0.6。组稀疏典型相关分析筛选出的特征的相关系数为0.626 9±0.016 1，稀疏典型相关系数为0.625 5±0.018 1。经过10次实验，在采用组稀疏典型相关分析方法筛选出的最相关的前20组特征中，属于已知的精神分裂症相关75个基因的比例大于随机选出的非相关基因的比例。结论 通过组稀疏典型相关分析方法能够筛选出多组精神分裂症的相关基因和脑区，为今后对精神分裂症等复杂精神类疾病的研究提供了新的思路。 |
| 英文摘要: |
| Objective To explore the correlation between imaging data and genetic data of schizophrenia patients using imaging genetics method. Methods A group sparse canonical analysis method was proposed, group sparse constraints λ1‖u‖G and λ2‖v‖G were added to sparse canonical correlation analysis model to select features groups. Then, features within one group were selected by sparse constraints τ1‖u‖1 and τ2‖v‖1. The imaging genetics method based on group sparse canonical correlation analysis method was used to analyze the correlation between brain regions and genes of schizophrenia, and the stability and ability of this method to select biomarkers were also verified. Results Several pairs canonical brain regions and genes were identified. The left insula and gene AKT1 demonstrated the most significant correlation (r=0.653 8), and r value between right rectus and gene DAOA, MAGI2 were larger than 0.6. The correlation coefficients of selected features were 0.626 9±0.016 1 with group sparse canonical correlation analysis and 0.625 5±0.018 1 with sparse canonical correlation analysis. After 10 selections, the proportion of 75 genes related to schizophrenia was higher than that of non-related genes randomly selected in the most related 20 genes selected by group sparse canonical correlation analysis. Conclusion Several pairs canonical brain regions and genes can be identified by the group sparse canonical analysis method, which may provide a new way for the study of schizophrenia and other complex mental disorders. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
