健康与遗传心理学研究室知识库

精神疾病是复杂疾病，多种因素共同参与导致其发生、发展。研究表明遗传因素在其中起重要作用。近年积累的精神疾病遗传数据鉴别出大量易感基因，但己发现基因可解释的遗传度仍然非常有限，更多微效和低频变异(罕见变异)尚待挖掘。我们函需用合适的策略深入挖掘更多精神疾病易感基因从而探索其遗传基础。因此，我们展开了以下三方面研究。

第一，方法论研究。不同精神疾病研究现状不同，为此我们提出了两种不同的整合挖掘策略。针对遗传信息分散于各异质性强的研究现状，我们提出了“基于文献来源的精神疾病遗传数据整合与挖掘策略”。为发现更多与疾病表型相关的变异，我们提出了“基于全基因组关联研究挖掘精神疾病新变异策略”。我们分别建立了两种挖掘策略的对应分析流程。策略一，整合与扩展分析己有文献变异，提供更多可靠候选，并且找到微效变异的共同作用途径;策略二为发现新变异(包括常见和罕见变异)提供可能性。这两种挖掘策略为后续挖掘不同精神疾病易感基因及其遗传机制打下坚实基础。

第二，利用文献来源的遗传数据整合与挖掘策略探索创伤后应激障碍(Posttraumatic stress disorder, PTSD)的遗传基础。我们初筛了1,762篇文献，系统、全面整合105篇PTSD遗传研究，得到核心数据集，提供目前遗传证据最多的PTSD相关基因清单;基于核心数据集进一步深入挖掘PTSD，通过蛋白相互作用(Protein-protein interaction, PPI)分析扩展得到潜在候选基因，如AKTI , UBC等;同时，通路富集分析结果佐证HPA轴在PTSD假说中的重要作用。此外，我们较为全面地基于文献数据展示了PTSD的遗传研究现状，为后续的研究提供了广泛而可靠的数据集，并且客观解释当前研究异质性问题。以上内容均整合在数据库PTSDgene (http://ptsdgene.psych.ac.cn/)中，供研究者自由查询使用。作为业界首个PTSD遗传学数据库，PTSDgene自上线以来受到广泛关注，月访问量高达2,460个用户。

第三，基于全基因组关联研究挖掘注意缺陷多动障碍(Attention deficity peractivity disorder, ADHD)这一工作记忆内表型的遗传基础。该研究为968例 ADHD患者，我们从变异位点/基因、通路水平和生物网络多个层面探索ADHD 工作记忆的遗传基础。罕见变异基因层面(SKAT-O)分析得到显著结果并且通过了多重校验(P <2. 8e-6，其中基因”7X2-CTNNDI与顺序数字记忆广度(digit span forward, DSDF显著相关(P=4.28e-08。己有文献证据表明该基因和精神疾病相关，基因表达分析显示TMX2在小脑半球、前额叶、大脑皮层等组织中的表达量相对其他组织较高。常见变异通路富集分析(FDR<0.05)揭示了神经突发育、神经元产生和发育等通路在ADHD中的重要作用。网络分析进一步挖掘出与CTNNDI直接相互作用的基因(如FYN等)在精神疾病相关表型中的重要作用。本研究提供了与工作记忆相关的证据，为后续进一步揭示ADHD的遗传基础提供线索。

综上所述，本论文基于精神疾病研究现状与挑战提出对己有变异的整合与挖掘策略、发现新变异的挖掘策略，并分别应用于PTSD和ADHD探索其遗传机制，己获得研究结果为后续实验验证和临床应用提供了重要线索。本研究的发现扩展了现有精神疾病遗传学研究的技术和方法，有助于进一步推动精神疾病的遗传机制研究。期望该挖掘策略不局限于上述两病，后续可应用于更多精神疾病。

Mental disorders is complex diseases, many factors contribute to its occurrence and development. Studies have shown that genetic factors play an important role. Recently, lots of genetic data on mental disorders have been accumulated, and large number of susceptibility genes have been identified. However, the heritability explained by the identified genes is still very limited. More minor and low frequency variants (rare variant) have yet to be discovered. It is urgent to use appropriate strategies to mine additional susceptibility genes for mental disorders and explore their genetic basis. Therefore, we have carried out the following three aspects of research.

First part of this thesis, methodology research. Different mental disorders research status is different, for which we propose two different integrated mining strategies. Since genetic information is scattered among various heterogeneous studies, we propose a "integration and mining strategy for genetic data based on literature sources". In order to discover more genes related to mental disorders, we propose another strategy "for mining new variants in mental disorders based on genome-wide association studies". We have established two analysis process for these mining strategies. The first strategy integrates and extends the existing literature variations to provide more reliable candidates and to find common approach for minor variants. The establishment of the second strategy provides us with the possibility to discover new variants, both common and rare. The research of these two mining strategies has laid a solid foundation for us to explore the susceptibility genes and their genetic basis of different mental disorders.

Second part of this thesis, the genetic basis of post-traumatic stress disorder (PTSD) was explored using genetic data integration and mining strategies from literature. We first screened 1,762 studies, and comprehensively integrated 105 PTSD genetic studies, finally obtained the core dataset and provided a list of PTSD-related genes with the most genetic evidence. Based on the core dataset, the PTSD was further explored, and potential candidate genes such as AKTI and UBC were obtained throuth protein-protein interaction (Protein-protein interaction, PPI) analysis. Meanwhile, the results of pathway enrichment analysis demonstrate the important role of HPA axis in PTSD hypothesis. In addition, we present the comprehensive status of genetic studies of PTSD based on literature data, providing a wide range and reliable data set for subsequent research, and objectively explaining the heterogeneity problem in current research. The above content is integrated in the database PTSDgene (http://ptsdgene.psych.ac.cn/) for researchers to use freely. As the first PTSD genetics database, PTSDgene has received extensive attention since its launch, with a monthly visit volume of 2,460 users.

Third part of this thesis, we explored the genetic basis of the endophenotype of working memory in attention deficit hyperactivity disorder (ADHD) based on genome-wide association studies. The study sample was 968 patients with ADHD. We explored the genetic basis of working memory in ADHD from multiple levels of variants/gene, pathway, and biological networks. The rare variant gene level (SKAT-O) analysis yielded significant results and passed multiple test (P<2.8e-6), in which the gene TMX2-CTNNDI (P=4.28e-08) was significantly associated with digit span forward (DSDF). Studies show that this gene is associated with mental disorders. Gene expression analysis shows that TMX2 is expressed higher in cerebellar hemisphere, prefrontal cortex, cerebral cortex than other tissues. Pathway enrichment analysis of common variants (FDR<0.05) revealed important roles in neurite development, neuronal production, and development in ADHD. Network analysis further explores the important role of genes that interact directly with CTNNDI (such as FYN, etc.) in the mental disorders associated phenotype. This study provides evidence related to working memory and provide clues for further revealing the genetic basis of ADHD.

In summary, we propose the integration and mining strategy for literatures and the genome-wide mining strategy for discovering new variants based on the current research status and challenges of mental disorders. Then we applied the strategy in PTSD and ADHD respectively to explore their genetic basis. The results we obtained provide important cules for subsequent experimental verification and clinical application. The findings of this study extend the existing techniques and methods of genetic research in mental disorders and help to further advance the genetic mechanisms of mental disorders. It is expected that the mining strategy is not limited to the diseases in this study and can be tried and applied in the subsequent mining of more diseases.