2/5 Bipolar-Schizophrenia Network for Intermediate Phenotypes 2 (B-SNIP2)

描述（由申请人提供）：主要精神病（SZ， SAD， BDP），当仅由临床现象定义时，在神经生物学，生物标志物，共病，症状和遗传特征上有广泛的重叠。当在疾病定义和识别中考虑生物变量时，我们的领域可能受益于在其他医学领域看到的疾病的转型重新概念化。精神病学中的这种方法将取决于：(i)使用定义明确的疾病域，（ii）捕获临床异质性并支持统计方法的大样本，以及（iii）获得可量化的实验室测量以重新概念化疾病特征的能力。5位点B-SNIP的焦点是精神病，这是一种理想的临床表型。B- SNIP1招募了2500多名志愿者，并在多个分析水平（认知、心理生理、脑成像、社会和临床）上进行了密集的表型分析。总体数据描述了DSM精神病诊断（BDP， SAD， SZ）中连续的表型改变，几乎没有诊断特异性的证据。为了尝试使用这些密集的表型特征来定义基于生物学的亚群，我们使用生物标志物和多阶段多变量分析程序对先证进行了重新分组。我们根据核心表型特征确定了3种精神病“生物型”。生物型在外部验证器之间显示出独特的差异，这些验证器在分类的初始构建中没有使用。B- snip2将使用增强的先证者数量、生物标志物面板和复杂的多元统计方法复制和扩展B- SNIP1。我们将在两个具体目标的范围内实现我们的目标。(1)建立“精神病生物标志物数据库”（PBD）：新招募3000名精神病先证者和600名健康志愿者，收集临床、心理社会、电生理、眼动、影像学和血液生物标志物等数据。核心生物标志物（用于生物型定义）和外部验证器（用于验证生物型的神经生物学独特性）将按规定收集。参与者的遗传特征将与布罗德研究所合作获得。(2)对比和测试精神病分类方法：基于PBD中的数据，评估不同分类结构定义精神病亚群的能力：(i) DSM, （ii）基于临床变量的B-SNIP2生物型，（iii） B-SNIP1生物型，（iv）基于生物标志物的B-SNIP2生成的生物型，以及(v)基于临床变量和生物标志物的B-SNIP2生物型。从传统的DSM诊断标准作为分类和测试(i)-(v)开始，我们将使用应用于外部生物标志物验证器的线性、二次和非参数判别函数分析来检查传统诊断系统与生物衍生分类（成像、社会心理和遗传外部验证器）之间的关联。我们将能够根据生物学特征确定最有力的分类方法。我们寻求一种合理的精神障碍分类，这将成功地识别新的疾病

DESCRIPTION (provided by applicant): The major psychoses (SZ, SAD, BDP), when defined by clinical phenomenology alone, overlap extensively on neurobiological, biomarker, co-morbid, symptomatic, and genetic characteristics. Our field may benefit from transformational re-conceptualizations of disease seen in other areas of medicine when biological variables are considered in disease definitions and identification. This approach in psychiatry will depend on: (i) use of well- defined disease domains, (ii) large samples that capture clinical heterogeneity and support statistical approaches, and (iii) ability to acquire quantifiable laboratory measures t inform re-conceptualization of disease characteristics. The 5-site B-SNIP focus is psychosis, an ideal clinical phenotype for this purpose. B- SNIP1 recruited over 2500 volunteers and performed dense phenotyping across multiple levels of analysis (cognitive, psychophysiological, brain imaging, social and clinical). The overall data described a continuum of phenotypic alterations across the DSM psychosis diagnoses (BDP, SAD, SZ) with little evidence of diagnostic specificity. In an attempt to use these dense phenotypic characteristics to define biologically based subgroups, we re-grouped probands using biomarkers and a multistage multivariate analysis procedure. We identified 3 psychosis "Biotypes" based on core phenotypic features. Biotypes showed unique differences across external validators that were not used in the initial construction of the categories. B-SNIP2 will replicate and extend B- SNIP1 using enhanced proband number, biomarker panel, and sophistication of multivariate statistical approaches. We will accomplish our goals within the context of two specific aims. SA(1) Construct a 'Psychosis Biomarker Database' (PBD): Recruit 3000 new psychosis probands and 600 healthy volunteers and collect data including clinical, psychosocial, electrophysiological, ocular motor, imaging and blood biomarkers. Core biomarkers (used for Biotype definition) and external validators (used for verifying neurobiological distinctiveness of Biotypes) will be collected as specified. Genetic characteristics of the participants will be obtained in collaboratin with the Broad Institute. SA(2) Contrast and test taxometric approaches to categorizing psychosis: Evaluate the ability of different taxonomic structures to define psychosis subgroups, based on data in the PBD: (i) DSM, (ii) B-SNIP2 biotypes based on clinical variables, (iii) B-SNIP1 Biotypes, (iv) B-SNIP2-generated biotypes based on biomarkers, and (v) B-SNIP2 biotypes based on both clinical variables and biomarkers. Beginning with traditional DSM diagnostic criteria as the taxonomy and testing (i)-(v) we will use linear, quadratic and nonparametric discriminant function analysis applied to external biomarker validators to examine the association between the traditional diagnostic system and the biologically- derived classification (imaging, psychosocial and genetic external validators). We will be able to determine the strongest taxonomic approach based on biological characteristics. We seek a rational classification of psychotic disorders that will be successful in identifying novel disease targets and treatments approaches.