Brain Connectome Mapping of Complex Human Traits and Their Polygenic Architecture Using Machine Learning

Polygenic architecture of brain structure and function, behaviors, and psychopathologies in children

The human brain undergoes structural and functional changes during childhood, a critical period in cognitive and behavioral development. Understanding the genetic architecture of the brain development in children can offer valuable insights into the development of the brain, cognition, and behaviors. Here, we integrated brain imaging-genetic-phenotype data from over 8,600 preadolescent children of diverse ethnic backgrounds using multivariate statistical techniques. We found a low-to-moderate level of SNP-based heritability in most IDPs, which is lower compared to the adult brain. Using sparse generalized canonical correlation analysis (SGCCA), we identified several covariation patterns among genome-wide polygenic scores (GPSs) of 29 traits, 7 different modalities of brain imaging-derived phenotypes (IDPs), and 266 cognitive and psychological phenotype data. In structural MRI, significant positive associations were observed between total grey matter volume, left ventral diencephalon volume, surface area of right accumbens and the GPSs of cognition-related traits. Conversely, negative associations were found with the GPSs of ADHD, depression and neuroticism. Additionally, we identified a significant positive association between educational attainment GPS and regional brain activation during the N-back task. The BMI GPS showed a positive association with fractional anisotropy (FA) of connectivity between the cerebellum cortex and amygdala in diffusion MRI, while the GPSs for educational attainment and cannabis use were negatively associated with the same IDPs. Our GPS-based prediction models revealed substantial genetic contributions to cognitive variability, while the genetic basis for many mental and behavioral phenotypes remained elusive. This study delivers a comprehensive map of the relationships between genetic profiles, neuroanatomical diversity, and the spectrum of cognitive and behavioral traits in preadolescence.

Linking Polygenic Risk of Schizophrenia to Variation in Magnetic Resonance Imaging Brain Measures: A Comprehensive Systematic Review

BACKGROUND AND HYPOTHESIS

Schizophrenia is highly heritable, with a polygenic effect of many genes conferring risk. Evidence on whether cumulative risk also predicts alterations in brain morphology and function is inconsistent. This systematic review examined evidence for schizophrenia polygenic risk score (sczPRS) associations with commonly used magnetic resonance imaging (MRI) measures. We expected consistent evidence to emerge for significant sczPRS associations with variation in structure and function, specifically in frontal, temporal, and insula cortices that are commonly implicated in schizophrenia pathophysiology.

STUDY DESIGN

In accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we searched MEDLINE, Embase, and PsycINFO for peer-reviewed studies published between January 2013 and March 2022. Studies were screened against predetermined criteria and National Institutes of Health (NIH) quality assessment tools.

STUDY RESULTS

In total, 57 studies of T1-weighted structural, diffusion, and functional MRI were included (age range = 9-80 years, Nrange = 64-76 644). We observed moderate, albeit preliminary, evidence for higher sczPRS predicting global reductions in cortical thickness and widespread variation in functional connectivity, and to a lesser extent, region-specific reductions in frontal and temporal volume and thickness. Conversely, sczPRS does not predict whole-brain surface area or gray/white matter volume. Limited evidence emerged for sczPRS associations with diffusion tensor measures of white matter microstructure in a large community sample and smaller cohorts of children and young adults. These findings were broadly consistent across community and clinical populations.

CONCLUSIONS

Our review supports the hypothesis that schizophrenia is a disorder of disrupted within and between-region brain connectivity, and points to specific whole-brain and regional MRI metrics that may provide useful intermediate phenotypes.

The overlapping genetic architecture of psychiatric disorders and cortical brain structure

Both psychiatric vulnerability and cortical structure are shaped by the cumulative effect of common genetic variants across the genome. However, the shared genetic underpinnings between psychiatric disorders and brain structural phenotypes, such as thickness and surface area of the cerebral cortex, remains elusive. In this study, we employed pleiotropy-informed conjunctional false discovery rate analysis to investigate shared loci across genome-wide association scans of regional cortical thickness, surface area, and seven psychiatric disorders in approximately 700,000 individuals of European ancestry. Aggregating regional measures, we identified 50 genetic loci shared between psychiatric disorders and surface area, as well as 26 genetic loci shared with cortical thickness. Risk alleles exhibited bidirectional effects on both cortical thickness and surface area, such that some risk alleles for each disorder increased regional brain size while other risk alleles decreased regional brain size. Due to bidirectional effects, in many cases we observed extensive pleiotropy between an imaging phenotype and a psychiatric disorder even in the absence of a significant genetic correlation between them. The impact of genetic risk for psychiatric disorders on regional brain structure did exhibit a consistent pattern across highly comorbid psychiatric disorders, with 80% of the genetic loci shared across multiple disorders displaying consistent directions of effect. Cortical patterning of genetic overlap revealed a hierarchical genetic architecture, with the association cortex and sensorimotor cortex representing two extremes of shared genetic influence on psychiatric disorders and brain structural variation. Integrating multi-scale functional annotations and transcriptomic profiles, we observed that shared genetic loci were enriched in active genomic regions, converged on neurobiological and metabolic pathways, and showed differential expression in postmortem brain tissue from individuals with psychiatric disorders. Cumulatively, these findings provide a significant advance in our understanding of the overlapping polygenic architecture between psychopathology and cortical brain structure.

A functional connectome signature of blood pressure in >30 000 participants from the UK biobank

AIMS

Elevated blood pressure (BP) is a prevalent modifiable risk factor for cardiovascular diseases and contributes to cognitive decline in late life. Despite the fact that functional changes may precede irreversible structural damage and emerge in an ongoing manner, studies have been predominantly informed by brain structure and group-level inferences. Here, we aim to delineate neurobiological correlates of BP at an individual level using machine learning and functional connectivity.

METHODS AND RESULTS

Based on whole-brain functional connectivity from the UK Biobank, we built a machine learning model to identify neural representations for individuals' past (∼8.9 years before scanning, N = 35 882), current (N = 31 367), and future (∼2.4 years follow-up, N = 3 138) BP levels within a repeated cross-validation framework. We examined the impact of multiple potential covariates, as well as assessed these models' generalizability across various contexts.The predictive models achieved significant correlations between predicted and actual systolic/diastolic BP and pulse pressure while controlling for multiple confounders. Predictions for participants not on antihypertensive medication were more accurate than for currently medicated patients. Moreover, the models demonstrated robust generalizability across contexts in terms of ethnicities, imaging centres, medication status, participant visits, gender, age, and body mass index. The identified connectivity patterns primarily involved the cerebellum, prefrontal, anterior insula, anterior cingulate cortex, supramarginal gyrus, and precuneus, which are key regions of the central autonomic network, and involved in cognition processing and susceptible to neurodegeneration in Alzheimer's disease. Results also showed more involvement of default mode and frontoparietal networks in predicting future BP levels and in medicated participants.

CONCLUSION

This study, based on the largest neuroimaging sample currently available and using machine learning, identifies brain signatures underlying BP, providing evidence for meaningful BP-associated neural representations in connectivity profiles.

Data sharing to advance gene-targeted therapies in rare diseases

Recent advancements in gene-targeted therapies have highlighted the critical role data sharing plays in successful translational drug development for people with rare diseases. To scale these efforts, we need to systematize these sharing principles, creating opportunities for more rapid, efficient, and scalable drug discovery/testing including long-term and transparent assessment of clinical safety and efficacy. A number of challenges will need to be addressed, including the logistical difficulties of studying rare diseases affecting individuals who may be scattered across the globe, scientific, technical, regulatory, and ethical complexities of data collection, and harmonization and integration across multiple platforms and contexts. The NCATS/NIH Gene-Targeted Therapies: Early Diagnosis and Equitable Delivery meeting series held during June 2021 included data sharing models that address these issues and framed discussions of areas that require improvement. This article describes these discussions and provides a series of considerations for future data sharing.

Cross-cohort replicability and generalizability of connectivity-based psychometric prediction patterns

An increasing number of studies have investigated the relationships between inter-individual variability in brain regions' connectivity and behavioral phenotypes, making use of large population neuroimaging datasets. However, the replicability of brain-behavior associations identified by these approaches remains an open question. In this study, we examined the cross-dataset replicability of brain-behavior association patterns for fluid cognition and openness predictions using a previously developed region-wise approach, as well as using a standard whole-brain approach. Overall, we found moderate similarity in patterns for fluid cognition predictions across cohorts, especially in the Human Connectome Project Young Adult, Human Connectome Project Aging, and Enhanced Nathan Kline Institute Rockland Sample cohorts, but low similarity in patterns for openness predictions. In addition, we assessed the generalizability of prediction models in cross-dataset predictions, by training the model in one dataset and testing in another. Making use of the region-wise prediction approach, we showed that first, a moderate extent of generalizability could be achieved with fluid cognition prediction, and that, second, a set of common brain regions related to fluid cognition across cohorts could be identified. Nevertheless, the moderate replicability and generalizability could only be achieved in specific contexts. Thus, we argue that replicability and generalizability in connectivity-based prediction remain limited and deserve greater attention in future studies.

Interpreting Brain Biomarkers: Challenges and solutions in interpreting machine learning-based predictive neuroimaging

Predictive modeling of neuroimaging data (predictive neuroimaging) for evaluating individual differences in various behavioral phenotypes and clinical outcomes is of growing interest. However, the field is experiencing challenges regarding the interpretability of the results. Approaches to defining the specific contribution of functional connections, regions, or networks in prediction models are urgently needed, which may help explore the underlying mechanisms. In this article, we systematically review the methods and applications for interpreting brain signatures derived from predictive neuroimaging based on a survey of 326 research articles. Strengths, limitations, and the suitable conditions for major interpretation strategies are also deliberated. In-depth discussion of common issues in existing literature and the corresponding recommendations to address these pitfalls are provided. We highly recommend exhaustive validation on the reliability and interpretability of the biomarkers across multiple datasets and contexts, which thereby could translate technical advances in neuroimaging into concrete improvements in precision medicine.

Shared and unique brain network features predict cognitive, personality, and mental health scores in the ABCD study

How individual differences in brain network organization track behavioral variability is a fundamental question in systems neuroscience. Recent work suggests that resting-state and task-state functional connectivity can predict specific traits at the individual level. However, most studies focus on single behavioral traits, thus not capturing broader relationships across behaviors. In a large sample of 1858 typically developing children from the Adolescent Brain Cognitive Development (ABCD) study, we show that predictive network features are distinct across the domains of cognitive performance, personality scores and mental health assessments. On the other hand, traits within each behavioral domain are predicted by similar network features. Predictive network features and models generalize to other behavioral measures within the same behavioral domain. Although tasks are known to modulate the functional connectome, predictive network features are similar between resting and task states. Overall, our findings reveal shared brain network features that account for individual variation within broad domains of behavior in childhood.

Mind the gap: Performance metric evaluation in brain-age prediction

Estimating age based on neuroimaging-derived data has become a popular approach to developing markers for brain integrity and health. While a variety of machine-learning algorithms can provide accurate predictions of age based on brain characteristics, there is significant variation in model accuracy reported across studies. We predicted age in two population-based datasets, and assessed the effects of age range, sample size and age-bias correction on the model performance metrics Pearson's correlation coefficient (r), the coefficient of determination (R² ), Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). The results showed that these metrics vary considerably depending on cohort age range; r and R² values are lower when measured in samples with a narrower age range. RMSE and MAE are also lower in samples with a narrower age range due to smaller errors/brain age delta values when predictions are closer to the mean age of the group. Across subsets with different age ranges, performance metrics improve with increasing sample size. Performance metrics further vary depending on prediction variance as well as mean age difference between training and test sets, and age-bias corrected metrics indicate high accuracy-also for models showing poor initial performance. In conclusion, performance metrics used for evaluating age prediction models depend on cohort and study-specific data characteristics, and cannot be directly compared across different studies. Since age-bias corrected metrics generally indicate high accuracy, even for poorly performing models, inspection of uncorrected model results provides important information about underlying model attributes such as prediction variance.

Brain age prediction using fMRI network coupling in youths and associations with psychiatric symptoms

OBJECTIVE

Magnetic resonance imaging (MRI) has shown that estimated brain age is deviant from chronological age in various common brain disorders. Brain age estimation could be useful for investigating patterns of brain maturation and integrity, aiding to elucidate brain mechanisms underlying these heterogeneous conditions. Here, we examined functional brain age in two large samples of children and adolescents and its relation to mental health.

METHODS

We used resting-state fMRI data from the Philadelphia Neurodevelopmental Cohort (PNC; n = 1126, age range 8-22 years) to estimate functional connectivity between brain networks, and utilized these as features for brain age prediction. We applied the prediction model to 1387 individuals (age range 8-22 years) in the Healthy Brain Network sample (HBN). In addition, we estimated brain age in PNC using a cross-validation framework. Next, we tested for associations between brain age gap and various aspects of psychopathology and cognitive performance.

RESULTS

Our model was able to predict age in the independent test samples, with a model performance of r = 0.54 for the HBN test set, supporting consistency in functional connectivity patterns between samples and scanners. Linear models revealed a significant association between brain age gap and psychopathology in PNC, where individuals with a lower estimated brain age, had a higher overall symptom burden. These associations were not replicated in HBN.

DISCUSSION

Our findings support the use of brain age prediction from fMRI-based connectivity. While requiring further extensions and validations, the approach may be instrumental for detecting brain phenotypes related to intrinsic connectivity and could assist in characterizing risk in non-typically developing populations.

Population modeling with machine learning can enhance measures of mental health

BACKGROUND

Biological aging is revealed by physical measures, e.g., DNA probes or brain scans. In contrast, individual differences in mental function are explained by psychological constructs, e.g., intelligence or neuroticism. These constructs are typically assessed by tailored neuropsychological tests that build on expert judgement and require careful interpretation. Could machine learning on large samples from the general population be used to build proxy measures of these constructs that do not require human intervention?

RESULTS

Here, we built proxy measures by applying machine learning on multimodal MR images and rich sociodemographic information from the largest biomedical cohort to date: the UK Biobank. Objective model comparisons revealed that all proxies captured the target constructs and were as useful, and sometimes more useful, than the original measures for characterizing real-world health behavior (sleep, exercise, tobacco, alcohol consumption). We observed this complementarity of proxy measures and original measures at capturing multiple health-related constructs when modeling from, both, brain signals and sociodemographic data.

CONCLUSION

Population modeling with machine learning can derive measures of mental health from heterogeneous inputs including brain signals and questionnaire data. This may complement or even substitute for psychometric assessments in clinical populations.

Sex-related differences in brain dynamism at rest as neural correlates of positive and negative valence system constructs

Clinical anxiety and depression are the most prevalent mental illnesses, likely representing maladaptive expressions of negative valence systems concerned with conditioned responses to fear, threat, loss, and frustrative nonreward. These conditions exhibit similar, striking sex/gender-related differences in onset, incidence, and severity for which the neural correlates are not yet established. In alarge sample of neurotypical young adults, we demonstrate that intrinsic brain dynamism metrics derived from sex-sensitive models of whole-brain network function are significantly associated with valence system traits. Surprisingly, we found that greater brain dynamism is strongly positively correlated to anxiety and depression traits in males, but almost wholly decoupled from traits for important cognitive control and reappraisal strategies associated with positive valence. Conversely, intrinsic brain dynamism is strongly positively coupled to drive, novelty-seeking and self-control in females with only rare or non-significant directional negative correlation with anxiety and depression traits. Our results suggest that the dynamic neural correlates of traits for valence, anxiety and depression are significantly different in males/men and females/women. These findings may relate to the known sex/gender-related differences in cognitive reappraisal of emotional experiences and clinical presentations of anxiety and depression, with potential relevance to gold standard therapies based on enhancing cognitive control strategies.

Functional Connectome Prediction of Anxiety Related to the COVID-19 Pandemic

OBJECTIVE

Increased anxiety in response to the COVID-19 pandemic has been widely noted. The purpose of this study was to test whether the prepandemic functional connectome predicted individual anxiety induced by the pandemic.

METHODS

Anxiety scores from healthy undergraduate students were collected during the severe and remission periods of the pandemic (first survey, February 22-28, 2020, N=589; second survey, April 24 to May 1, 2020, N=486). Brain imaging data and baseline (daily) anxiety ratings were acquired before the pandemic. The predictive performance of the functional connectome on individual anxiety was examined using machine learning and was validated in two external undergraduate student samples (N=149 and N=474). The clinical relevance of the findings was further explored by applying the connectome-based neuromarkers of pandemic-related anxiety to distinguish between individuals with specific mental disorders and matched healthy control subjects (generalized anxiety disorder, N=43; major depression, N=536; schizophrenia, N=72).

RESULTS

Anxiety scores increased from the prepandemic baseline to the severe stage of the pandemic and remained high in the remission stage. The prepandemic functional connectome predicted pandemic-related anxiety and generalized to the external sample but showed poor performance for predicting daily anxiety. The connectome-based neuromarkers of pandemic-related anxiety further distinguished between participants with generalized anxiety and healthy control subjects but were not useful for diagnostic classification in major depression and schizophrenia.

CONCLUSIONS

These findings demonstrate the feasibility of using the functional connectome to predict individual anxiety induced by major stressful events (e.g., the current global health crisis), which advances our understanding of the neurobiological basis of anxiety susceptibility and may have implications for developing targeted psychological and clinical interventions that promote the reduction of stress and anxiety.

Multimodal hippocampal and amygdala subfield volumetry in polygenic risk for Alzheimer's disease

Preclinical models of Alzheimer's disease (AD) suggest that volumetric reductions in medial temporal lobe (MTL) structures manifest before clinical onset. AD polygenic risk scores (PRSs) are further linked to reduced MTL volumes (the hippocampus/amygdala); however, the relationship between the PRS and specific subregions remains unclear. We determine the relationship between the AD-PRSs and MTL subregions in a large sample of young participants (N = 730, aged 22–35 years) using a multimodal (T1w/T2w) approach. We first demonstrate that the PRSs for the hippocampus/amygdala predict their respective volumes and specific hippocampal subregions (pFDR < 0.05). We further observe negative relationships between the AD-PRSs and whole hippocampal/amygdala volumes. Critically, we demonstrate novel associations between the AD-PRSs and specific hippocampal subfields such as CA1 (β = −0.096, pFDR = 0.045) and the fissure (β = −0.101, pFDR = 0.041). We provide evidence that the AD-PRS is linked to specific MTL subfields decades before AD onset. This may help inform preclinical models of AD risk, providing additional specificity for intervention and further insight into mechanisms by which common AD variants confer susceptibility.

•

Polygenic risk for Alzheimer's disease (AD-PRS) explains significant proportion of AD.

•

AD-PRS also linked to hippocampus and amygdala volume.

•

AD-PRS is negatively associated with specific hippocampal subfields.

•

Polygenic AD models help us understand genetic contributions to medial temporal lobe nuclei.

Differences in directed functional brain connectivity related to age, sex and mental health

Functional interconnections between brain regions define the "connectome" which is of central interest for understanding human brain function. Resting-state functional magnetic resonance (rsfMRI) work has revealed changes in static connectivity related to age, sex, cognitive abilities and psychiatric symptoms, yet little is known how these factors may alter the information flow. The commonly used approach infers functional brain connectivity using stationary coefficients yielding static estimates of the undirected connection strength between brain regions. Dynamic graphical models (DGMs) are a multivariate model with dynamic coefficients reflecting directed temporal associations between nodes, and can yield novel insight into directed functional connectivity. Here, we leveraged this approach to test for associations between edge-wise estimates of direction flow across the functional connectome and age, sex, intellectual abilities and mental health. We applied DGM to investigate patterns of information flow in data from 984 individuals from the Human Connectome Project (HCP) and 10,249 individuals from the UK Biobank. Our analysis yielded patterns of directed connectivity in independent HCP and UK Biobank data similar to those previously reported, including that the cerebellum consistently receives information from other networks. We show robust associations between information flow and age and sex for several connections, with strongest effects of age observed in the sensorimotor network. Visual, auditory and sensorimotor nodes were also linked to mental health. Our findings support the use of DGM as a measure of directed connectivity in rsfMRI data and provide new insight into the shaping of the connectome during aging.

Convergent molecular, cellular, and cortical neuroimaging signatures of major depressive disorder

Major depressive disorder is a debilitating condition with diverse neuroimaging correlates, including cortical thinning in medial prefrontal cortex and altered functional connectivity of cortical association networks. However, the molecular bases of these imaging markers remain ambiguous, despite a need for treatment targets and mechanisms. Here, we advance cross-modal approaches to identify cell types and gene transcripts associated with depression-implicated cortex. Across multiple population-imaging datasets (combined N ≥ 23,723) and ex vivo patient cortical tissue, somatostatin interneurons and astrocytes emerge as replicable cell-level correlates of depression and negative affect. These data identify transcripts, cell types, and molecular processes associated with neuroimaging markers of depression and offer a roadmap for integrating in vivo clinical imaging with genetic and postmortem patient transcriptional data.

Major depressive disorder emerges from the complex interactions of biological systems that span genes and molecules through cells, networks, and behavior. Establishing how neurobiological processes coalesce to contribute to depression requires a multiscale approach, encompassing measures of brain structure and function as well as genetic and cell-specific transcriptional data. Here, we examine anatomical (cortical thickness) and functional (functional variability, global brain connectivity) correlates of depression and negative affect across three population-imaging datasets: UK Biobank, Brain Genomics Superstruct Project, and Enhancing NeuroImaging through Meta Analysis (ENIGMA; combined n ≥ 23,723). Integrative analyses incorporate measures of cortical gene expression, postmortem patient transcriptional data, depression genome-wide association study (GWAS), and single-cell gene transcription. Neuroimaging correlates of depression and negative affect were consistent across three independent datasets. Linking ex vivo gene down-regulation with in vivo neuroimaging, we find that transcriptional correlates of depression imaging phenotypes track gene down-regulation in postmortem cortical samples of patients with depression. Integrated analysis of single-cell and Allen Human Brain Atlas expression data reveal somatostatin interneurons and astrocytes to be consistent cell associates of depression, through both in vivo imaging and ex vivo cortical gene dysregulation. Providing converging evidence for these observations, GWAS-derived polygenic risk for depression was enriched for genes expressed in interneurons, but not glia. Underscoring the translational potential of multiscale approaches, the transcriptional correlates of depression-linked brain function and structure were enriched for disorder-relevant molecular pathways. These findings bridge levels to connect specific genes, cell classes, and biological pathways to in vivo imaging correlates of depression.

A principal component approach to improve association testing with polygenic risk scores

Polygenic risk scores (PRSs) have become an increasingly popular approach for demonstrating polygenic influences on complex traits and for establishing common polygenic signals between different traits. PRSs are typically constructed using pruning and thresholding (P+T), but the best choice of parameters is uncertain; thus multiple settings are used and the best is chosen. Optimization can lead to inflated Type I error. Permutation procedures can correct this, but they can be computationally intensive. Alternatively, a single parameter setting can be chosen a priori for the PRS, but choosing suboptimal settings results in loss of power. We propose computing PRSs under a range of parameter settings, performing principal component analysis (PCA) on the resulting set of PRSs, and using the first PRS-PC in association tests. The first PC reweights the variants included in the PRS to achieve maximum variation over all PRS settings used. Using simulations and a real data application to study PRS association with bipolar disorder and psychosis in bipolar disorder, we compare the performance of the proposed PRS-PCA approach with a permutation test and an a priori selected p-value threshold. The PRS-PCA approach is simple to implement, outperforms the other strategies in most scenarios, and provides an unbiased estimate of prediction performance.