Leveraging computational strategies to disentangle the genetic and neural underpinnings of ADHD and its associated cognitive systems

利用计算策略来解开 ADHD 及其相关认知系统的遗传和神经基础

• 批准号: 10732355
• 负责人: Tian Ge
• 金额: $ 76.88万
• 依托单位: MASSACHUSETTS GENERAL HOSPITAL
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-09-08 至 2028-07-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10732355
• 关键词: Address; Adolescent; Affect; Atlases; Attention deficit hyperactivity disorder; Automobile Driving; Behavior; Behavioral Symptoms; Biological; Brain; Brain imaging; Brain region; Child; Child Psychiatry; Clinic; Clinical Psychology; Cognition; Cognitive; Collaborations; Complex; Data; Diagnosis; Diagnostic; Dimensions; Disease; Ethnic Origin; Etiology; Executive Dysfunction; Family Study; Future; Genes; Genetic; Genetic Models; Genetic Risk; Genetic Variation; Genomics; Goals; Heritability; Hyperactivity; Image; Impairment; Impulsivity; Investigation; Knowledge; Link; MRI Scans; Machine Learning; Methods; Modeling; National Institute of Mental Health; Neuroanatomy; Neurobiology; Neurocognition; Outcome; Patient Care; Pharmacological Treatment; Phenotype; Psychopathology; Public Health; Research; Research Domain Criteria; Resources; Risk; Sampling; Statistical Models; Symptoms; System; Taxonomy; Therapeutic; Time; Training; Twin Studies; Youth; age group; base; behavioral phenotyping; biobank; cognitive development; cognitive function; cognitive system; computer studies; deep learning; driving behavior; executive function; functional outcomes; genomic variation; improved; inattention; lens; machine learning model; member; neural; neural circuit; neuroimaging; neuroimaging marker; neuropsychiatry; novel; novel therapeutics; psychosocial; racial diversity; response; risk stratification; serial imaging; trait; treatment response; treatment stratification

Impaired higher-order cognition is well-documented in different forms of neuropsychiatric illness, present in unaffected relatives, often presumed to underlie behavioral symptoms, and associated with functional outcome. Cognition is thus exceedingly relevant to studies of the etiology and trajectory of psychopathology. Nonetheless, the relationship between cognitive decrements and specific psychopathological conditions is not yet resolved. There are particular knowledge gaps in this regard for attention deficit/hyperactivity disorder (ADHD), one of the most common child psychiatry conditions worldwide. While models of ADHD have long highlighted executive functions (EF) as driving the behavioral symptoms of the condition, cognition in ADHD is increasingly acknowledged to be complex. Twin and family studies link aspects of cognition to ADHD risk, but findings are inconsistent. Also, not all affected youth show EF deficits, and domains separable from EF are impaired to varying degrees. Moreover, cognitive decrements within and beyond ADHD disrupt academic and psychosocial functioning and show limited response to pharmacologic treatments that benefit the disorder. Thus, understanding the overlapping and separable heritable neurobiology of ADHD and its related cognitive systems has implications for psychopathology models and patient care. In this proposal, we will study these issues through the lens of the NIMH’s Research Domain Criteria (RDoC) framework, which encourages a dimensional approach, interrogation of specific transdiagnostic traits, and multi-level links across genetics, brain systems and behavior. We will also capitalize on new resources in the field, i.e., advances in cognition genomics, novel computational strategies, and the Adolescent Brain Cognitive Development (ABCD) study. Our goal is to validate and demarcate the heritable biological underpinnings of specific cognitive systems that overlap with and extend beyond the ADHD construct. Our aims converge with NIMH PAR- 21-263, which seeks computational studies to validate dimensional constructs represented in the RDoC matrix in relation to psychopathology. Aim 1 will extract the genetic basis of ADHD-related cognitive constructs represented in RDoC using samples in the UK Biobank and comprehensively characterize their overlap with the genetic risk for ADHD using cutting-edge statistical genetics methods. Aim 2 will develop and validate an efficient, deep learning based, longitudinal neuroimaging processing pipeline that can extend to youth with diverse racial and ethnic backgrounds, and produce an atlas of brain regions and neural circuitry that underlie the cognitive constructs and the ADHD dimension. Aim 3 will train machine learning models that can integrate genetic risk scores, neuroimaging markers, and cognitive and behavioral phenotypes to predict future ADHD symptoms, cognitive functioning and functional outcomes. Completing these aims will advance models of the overlapping and distinct neurobiological bases of ADHD and its associated cognitive systems, which, in turn, will yield opportunities for risk stratification and novel therapeutics.

高阶认知受损在不同形式的神经精神疾病中有充分的证据，存在于