Comparison of variants of canonical correlation analysis and partial least squares for combined analysis of MRI and genetic data

A Novel Event-Related Neural Activity Extraction Technique by Fusing Neural Oscillation and Neural Synchronization

Previous research has proposed a number of techniques for the extraction of single-trial event-related neural activity (ERNA). However, these single-trial extraction techniques did not simultaneously consider about neural oscillation and neural synchronization, thereby creating an opportunity to optimize the single-trial extraction technique. In this study, based on the neural basis of electroencephalography (EEG), a novel single-trial extraction technique, which fuses neural oscillation and neural synchronization, was proposed. The neural activity extracted by this technique was identified as a novel type of ERNA and defined as connectivity-related neural activity (CRNA). The CRNA from the single-trial EEG was extracted through the utilisation of independent component analysis (ICA), non-negative matrix factorization (NNMF), and the alternating direction method of multipliers (ADMM). Furthermore, the performance of CRNA was evaluated using two key measures : signal-to-noise ratio (SNR) and Pearson's correlation coefficient (RHO). To investigate the effect of electrode density on CRNA performance, we compared five standard electrode configurations (including 9, 16, 32, 64, and 128 channel configurations). It was observed that there was a notable enhancement in the SNR of CRNA with an increase in electrode density. Finally, the comparative analysis demonstrated that the SNR and RHO of CRNA exhibited superior performance compared to several existing single-trial extraction techniques, including the conventional wavelet-based approach, the translation-invariant (TI) wavelet-based approach, the NZT, and the Generalized Subspace Approach (GSA). The proposed method facilitated the optimization of the single-trial extraction technique.

FPLS-DC: functional partial least squares through distance covariance for imaging genetics

MOTIVATION

Imaging genetics integrates imaging and genetic techniques to examine how genetic variations influence the function and structure of organs like the brain or heart, providing insights into their impact on behavior and disease phenotypes. The use of organ-wide imaging endophenotypes has increasingly been used to identify potential genes associated with complex disorders. However, analyzing organ-wide imaging data alongside genetic data presents two significant challenges: high dimensionality and complex relationships. To address these challenges, we propose a novel, nonlinear inference framework designed to partially mitigate these issues.

RESULTS

We propose a functional partial least squares through distance covariance (FPLS-DC) framework for efficient genome wide analyses of imaging phenotypes. It consists of two components. The first component utilizes the FPLS-derived base functions to reduce image dimensionality while screening genetic markers. The second component maximizes the distance correlation between genetic markers and projected imaging data, which is a linear combination of the FPLS-basis functions, using simulated annealing algorithm. In addition, we proposed an iterative FPLS-DC method based on FPLS-DC framework, which effectively overcomes the influence of inter-gene correlation on inference analysis. We efficiently approximate the null distribution of test statistics using a gamma approximation. Compared to existing methods, FPLS-DC offers computational and statistical efficiency for handling large-scale imaging genetics. In real-world applications, our method successfully detected genetic variants associated with the hippocampus, demonstrating its value as a statistical toolbox for imaging genetic studies.

AVAILABILITY AND IMPLEMENTATION

The FPLS-DC method we propose opens up new research avenues and offers valuable insights for analyzing functional and high-dimensional data. In addition, it serves as a useful tool for scientific analysis in practical applications within the field of imaging genetics research. The R package FPLS-DC is available in Github: https://github.com/BIG-S2/FPLSDC.

Identifying the joint signature of brain atrophy and gene variant scores in Alzheimer's Disease

The joint modeling of genetic data and brain imaging information allows for determining the pathophysiological pathways of neurodegenerative diseases such as Alzheimer's disease (AD). This task has typically been approached using mass-univariate methods that rely on a complete set of Single Nucleotide Polymorphisms (SNPs) to assess their association with selected image-derived phenotypes (IDPs). However, such methods are prone to multiple comparisons bias and, most importantly, fail to account for potential cross-feature interactions, resulting in insufficient detection of significant associations. Ways to overcome these limitations while reducing the number of traits aim at conveying genetic information at the gene level and capturing the integrated genetic effects of a set of genetic variants, rather than looking at each SNP individually. Their associations with brain IDPs are still largely unexplored in the current literature, though they can uncover new potential genetic determinants for brain modulations in the AD continuum. In this work, we explored an explainable multivariate model to analyze the genetic basis of the grey matter modulations, relying on the AD Neuroimaging Initiative (ADNI) phase 3 dataset. Cortical thicknesses and subcortical volumes derived from T1-weighted Magnetic Resonance were considered to describe the imaging phenotypes. At the same time the genetic counterpart was represented by gene variant scores extracted by the Sequence Kernel Association Test (SKAT) filtering model. Moreover, transcriptomic analysis was carried on to assess the expression of the resulting genes in the main brain structures as a form of validation. Results highlighted meaningful genotype-phenotype interactionsas defined by three latent components showing a significant difference in the projection scores between patients and controls. Among the significant associations, the model highlighted EPHX1 and BCAS1 gene variant scores involved in neurodegenerative and myelination processes, hence relevant for AD. In particular, the first was associated with decreased subcortical volumes and the second with decreasedtemporal lobe thickness. Noteworthy, BCAS1 is particularly expressed in the dentate gyrus. Overall, the proposed approach allowed capturing genotype-phenotype interactions in a restricted study cohort that was confirmed by transcriptomic analysis, offering insights into the underlying mechanisms of neurodegeneration in AD in line with previous findings and suggesting new potential disease biomarkers.

Cognition's dependence on functional network integrity with age is conditional on structural network integrity

Maintaining good cognitive function is crucial for well-being across the lifespan. We proposed that the degree of cognitive maintenance is determined by the functional interactions within and between large-scale brain networks. Such connectivity can be represented by the white matter architecture of structural brain networks that shape intrinsic neuronal activity into integrated and distributed functional networks. We explored how the function-structure connectivity convergence, and the divergence of functional connectivity from structural connectivity, contribute to the maintenance of cognitive function across the adult lifespan. Multivariate analyses were used to investigate the relationship between function-structure connectivity convergence and divergence with multivariate cognitive profiles, respectively. Cognitive function was increasingly dependent on function-structure connectivity convergence as age increased. The dependency of cognitive function on connectivity was particularly strong for high-order cortical networks and subcortical networks. The results suggest that brain functional network integrity sustains cognitive functions in old age, as a function of the integrity of the brain's structural connectivity.

Joint Sparse Collaborative Regression on Imaging Genetics Study of Schizophrenia

The imaging genetics approach generates large amount of high dimensional and multi-modal data, providing complementary information for comprehensive study of Schizophrenia, a complex mental disease. However, at the same time, the variety of these data in structures, resolutions, and formats makes their integrative study a forbidding task. In this paper, we propose a novel model called Joint Sparse Collaborative Regression (JSCoReg), which can extract class-specific features from different health conditions/disease classes. We first evaluate the performance of feature selection in terms of Receiver operating characteristic curve and the area under the ROC curve in the simulation experiment. We demonstrate that the JSCoReg model can achieve higher accuracy compared with similar models including Joint Sparse Canonical Correlation Analysis and Sparse Collaborative Regression. We then applied the JSCoReg model to the analysis of schizophrenia dataset collected from the Mind Clinical Imaging Consortium. The JSCoReg enables us to better identify biomarkers associated with schizophrenia, which are verified to be both biologically and statistically significant.

Microevolutionary variation in molar morphology of Onychomys leucogaster decoupled from genetic structure

In neutral models of quantitative trait evolution, both genetic and phenotypic divergence scale as random walks, producing a correlation between the two measures. However, complexity in the genotype-phenotype map may alter the correlation between genotypic and phenotypic divergence, even when both are evolving neutrally or nearly so. Understanding this correlation between phenotypic and genetic variation is critical for accurately interpreting the fossil record. This study compares the geographic structure and scaling of morphological variation of the shape of the first lower molar of 77 individuals of the northern grasshopper mouse Onychomys leucogaster to genome-wide SNP variation in the same sample. We found strong genetic structure but weak or absent morphological structure indicating that the scaling of each type of variation is decoupled from one another. Low P_ST values relative to F_ST values are consistent with a lack of morphological divergence in contrast to genetic divergence between groups. This lack of phenotypic structure and the presence of notable within-sample phenotypic variance are consistent with uniform selection or constraints on molar shape across a wide geographic and environmental range. Over time, this kind of decoupling may result in patterns of phenotypic stasis masking underlying genetic patterns.

Sparse Hierarchical Representation Learning on Functional Brain Networks for Prediction of Autism Severity Levels

Machine learning algorithms have been widely applied in diagnostic tools for autism spectrum disorder (ASD), revealing an altered brain connectivity. However, little is known about whether an magnetic resonance imaging (MRI)-based brain network is related to the severity of ASD symptoms in a large-scale cohort. We propose a graph convolution neural network-based framework that can generate sparse hierarchical graph representations for functional brain connectivity. Instead of assigning initial features for each node, we utilized a feature extractor to derive node features and the extracted representations can be fed to a hierarchical graph self-attention framework to effectively represent the entire graph. By incorporating connectivity embeddings in the feature extractor, we propose adjacency embedding networks to characterize the heterogeneous representations of the brain connectivity. Our proposed model variants outperform the benchmarking model with different configurations of adjacency embedding networks and types of functional connectivity matrices. Using this approach with the best configuration (SHEN atlas for node definition, Tikhonov correlation for connectivity estimation, and identity-adjacency embedding), we were able to predict individual ASD severity levels with a meaningful accuracy: the mean absolute error (MAE) and correlation between predicted and observed ASD severity scores resulted in 0.96, and r = 0.61 (P < 0.0001), respectively. To obtain a better understanding on how to generate better representations, we investigate the relationships between the extracted feature embeddings and the graph theory-based nodal measurements using canonical correlation analysis. Finally, we visualized the model to identify the most contributive functional connections for predicting ASD severity scores.

Improved Interpretability of Brain-Behavior CCA With Domain-Driven Dimension Reduction

Canonical Correlation Analysis (CCA) has been widely applied to study correlations between neuroimaging data and behavioral data. Practical use of CCA typically requires dimensionality reduction with, for example, Principal Components Analysis (PCA), however, this can result in CCA components that are difficult to interpret. In this paper, we introduce a Domain-driven Dimension Reduction (DDR) method, reducing the dimensionality of the original datasets and combining human knowledge of the structure of the variables studied. We apply the method to the Human Connectome Project S1200 release and compare standard PCA across all variables with DDR applied to individual classes of variables, finding that DDR-CCA results are more stable and interpretable, allowing the contribution of each class of variable to be better understood. By carefully designing the analysis pipeline and cross-validating the results, we offer more insights into the interpretation of CCA applied to brain-behavior data.

Identification of multimodal brain imaging association via a parameter decomposition based sparse multi-view canonical correlation analysis method

BACKGROUND

With the development of noninvasive imaging technology, collecting different imaging measurements of the same brain has become more and more easy. These multimodal imaging data carry complementary information of the same brain, with both specific and shared information being intertwined. Within these multimodal data, it is essential to discriminate the specific information from the shared information since it is of benefit to comprehensively characterize brain diseases. While most existing methods are unqualified, in this paper, we propose a parameter decomposition based sparse multi-view canonical correlation analysis (PDSMCCA) method. PDSMCCA could identify both modality-shared and -specific information of multimodal data, leading to an in-depth understanding of complex pathology of brain disease.

RESULTS

Compared with the SMCCA method, our method obtains higher correlation coefficients and better canonical weights on both synthetic data and real neuroimaging data. This indicates that, coupled with modality-shared and -specific feature selection, PDSMCCA improves the multi-view association identification and shows meaningful feature selection capability with desirable interpretation.

CONCLUSIONS

The novel PDSMCCA confirms that the parameter decomposition is a suitable strategy to identify both modality-shared and -specific imaging features. The multimodal association and the diverse information of multimodal imaging data enable us to better understand the brain disease such as Alzheimer's disease.

The "Neural Shift" of Sleep Quality and Cognitive Aging: A Resting-State MEG Study of Transient Neural Dynamics

Sleep quality changes dramatically from young to old age, but its effects on brain dynamics and cognitive functions are not yet fully understood. We tested the hypothesis that a shift in brain networks dynamics relates to sleep quality and cognitive performance across the lifespan. Network dynamics were assessed using Hidden Markov Models (HMMs) in resting-state MEG data from a large cohort of population-based adults (N = 564, aged 18-88). Using multivariate analyses of brain-sleep profiles and brain-cognition profiles, we found an age-related "neural shift," expressed as decreased occurrence of "lower-order" brain networks coupled with increased occurrence of "higher-order" networks. This "neural shift" was associated with both increased sleep dysfunction and decreased fluid intelligence, and this relationship was not explained by age, sex or other covariates. These results establish the link between poor sleep quality, as evident in aging, and a behavior-related shift in neural dynamics.

Differences in Performance of ASD and ADHD Subjects Facing Cognitive Loads in an Innovative Reasoning Experiment

We aim to investigate whether EEG dynamics differ in adults with ASD (Autism Spectrum Disorders) and ADHD (attention-deficit/hyperactivity disorder) compared with healthy subjects during the performance of an innovative cognitive task, Aristotle's valid and invalid syllogisms, and how these differences correlate with brain regions and behavioral data for each subject. We recorded EEGs from 14 scalp electrodes (channels) in 21 adults with ADHD, 21 with ASD, and 21 healthy, normal subjects. The subjects were exposed in a set of innovative cognitive tasks (inducing varying cognitive loads), Aristotle's two types of syllogism mentioned above. A set of 39 questions were given to participants related to valid-invalid syllogisms as well as a separate set of questionnaires, in order to collect a number of demographic and behavioral data, with the aim of detecting shared information with values of a feature extracted from EEG, the multiscale entropy (MSE), in the 14 channels ('brain regions'). MSE, a nonlinear information-theoretic measure of complexity, was computed to extract a feature that quantifies the complexity of the EEG. Behavior-Partial Least Squares Correlation, PLSC, is the method to detect the correlation between two sets of data, brain, and behavioral measures. -PLSC, a variant of PLSC, was applied to build a functional connectivity of the brain regions involved in the reasoning tasks. Graph-theoretic measures were used to quantify the complexity of the functional networks. Based on the results of the analysis described in this work, a mixed 14 × 2 × 3 ANOVA showed significant main effects of group factor and brain region* syllogism factor, as well as a significant brain region* group interaction. There are significant differences between the means of MSE (complexity) values at the 14 channels of the members of the 'pathological' groups of participants, i.e., between ASD and ADHD, while the difference in means of MSE between both ASD and ADHD and that of the control group is not significant. In conclusion, the valid-invalid type of syllogism generates significantly different complexity values, MSE, between ASD and ADHD. The complexity of activated brain regions of ASD participants increased significantly when switching from a valid to an invalid syllogism, indicating the need for more resources to 'face' the task escalating difficulty in ASD subjects. This increase is not so evident in both ADHD and control. Statistically significant differences were found also in the behavioral response of ASD and ADHD, compared with those of control subjects, based on the principal brain and behavior saliences extracted by PLSC. Specifically, two behavioral measures, the emotional state and the degree of confidence of participants in answering questions in Aristotle's valid-invalid syllogisms, and one demographic variable, age, statistically and significantly discriminate the three groups' ASD. The seed-PLC generated functional connectivity networks for ASD, ADHD, and control, were 'projected' on the regions of the Default Mode Network (DMN), the 'reference' connectivity, of which the structural changes were found significant in distinguishing the three groups. The contribution of this work lies in the examination of the relationship between brain activity and behavioral responses of healthy and 'pathological' participants in the case of cognitive reasoning of the type of Aristotle's valid and invalid syllogisms, using PLSC, a machine learning approach combined with MSE, a nonlinear method of extracting a feature based on EEGs that captures a broad spectrum of EEGs linear and nonlinear characteristics. The results seem promising in adopting this type of reasoning, in the future, after further enhancements and experimental tests, as a supplementary instrument towards examining the differences in brain activity and behavioral responses of ASD and ADHD patients. The application of the combination of these two methods, after further elaboration and testing as new and complementary to the existing ones, may be considered as a tool of analysis in helping detecting more effectively such types of disorders.

Do ADHD-impulsivity and BMI have shared polygenic and neural correlates?

There is an extensive body of literature linking ADHD to overweight and obesity. Research indicates that impulsivity features of ADHD account for a degree of this overlap. The neural and polygenic correlates of this association have not been thoroughly examined. In participants of the IMAGEN study, we found that impulsivity symptoms and body mass index (BMI) were associated (r = 0.10, n = 874, p = 0.014 FWE corrected), as were their respective polygenic risk scores (PRS) (r = 0.17, n = 874, p = 6.5 × 10-6 FWE corrected). We then examined whether the phenotypes of impulsivity and BMI, and the PRS scores of ADHD and BMI, shared common associations with whole-brain grey matter and the Monetary Incentive Delay fMRI task, which associates with reward-related impulsivity. A sparse partial least squared analysis (sPLS) revealed a shared neural substrate that associated with both the phenotypes and PRS scores. In a last step, we conducted a bias corrected bootstrapped mediation analysis with the neural substrate score from the sPLS as the mediator. The ADHD PRS associated with impulsivity symptoms (b = 0.006, 90% CIs = 0.001, 0.019) and BMI (b = 0.009, 90% CIs = 0.001, 0.025) via the neuroimaging substrate. The BMI PRS associated with BMI (b = 0.014, 95% CIs = 0.003, 0.033) and impulsivity symptoms (b = 0.009, 90% CIs = 0.001, 0.025) via the neuroimaging substrate. A common neural substrate may (in part) underpin shared genetic liability for ADHD and BMI and the manifestation of their (observable) phenotypic association.

Modeling brain, symptom, and behavior in the winds of change

Neuropsychopharmacology addresses pressing questions in the study of three intertwined complex systems: the brain, human behavior, and symptoms of illness. The field seeks to understand the perturbations that impinge upon those systems, either driving greater health or illness. In the pursuit of this aim, investigators often perform analyses that make certain assumptions about the nature of the systems that are being perturbed. Those assumptions can be encoded in powerful computational models that serve to bridge the wide gulf between a descriptive analysis and a formal theory of a system's response. Here we review a set of three such models along a continuum of complexity, moving from a local treatment to a network treatment: one commonly applied form of the general linear model, impulse response models, and network control models. For each, we describe the model's basic form, review its use in the field, and provide a frank assessment of its relative strengths and weaknesses. The discussion naturally motivates future efforts to interlink data analysis, computational modeling, and formal theory. Our goal is to inspire practitioners to consider the assumptions implicit in their analytical approach, align those assumptions to the complexity of the systems under study, and take advantage of exciting recent advances in modeling the relations between perturbations and system function.

A technical review of canonical correlation analysis for neuroscience applications

Collecting comprehensive data sets of the same subject has become a standard in neuroscience research and uncovering multivariate relationships among collected data sets have gained significant attentions in recent years. Canonical correlation analysis (CCA) is one of the powerful multivariate tools to jointly investigate relationships among multiple data sets, which can uncover disease or environmental effects in various modalities simultaneously and characterize changes during development, aging, and disease progressions comprehensively. In the past 10 years, despite an increasing number of studies have utilized CCA in multivariate analysis, simple conventional CCA dominates these applications. Multiple CCA-variant techniques have been proposed to improve the model performance; however, the complicated multivariate formulations and not well-known capabilities have delayed their wide applications. Therefore, in this study, a comprehensive review of CCA and its variant techniques is provided. Detailed technical formulation with analytical and numerical solutions, current applications in neuroscience research, and advantages and limitations of each CCA-related technique are discussed. Finally, a general guideline in how to select the most appropriate CCA-related technique based on the properties of available data sets and particularly targeted neuroscience questions is provided.

Multiscale neurobiological correlates of human neuroticism

Neuroticism is a heritable personality trait associated with negative emotionality; however, we know little regarding the association between the microscale and macroscale neurobiological substrates of human neuroticism. Cross‐scale correlation analysis may provide such information. In this study, voxel‐wise neuroimaging–neuroticism correlation analyses consistently showed a positive correlation between neuroticism and functional connectivity density (FCD) in the ventral striatum in 274 young Chinese adults. Partial least squares regression analysis showed that the FCD‐neuroticism correlation map was significantly spatially correlated with gene expression profiles in each of six donated human brains. Neuroticism‐related genes derived from the six donors consistently showed significant enrichment in the chemical synaptic transmission, circadian entrainment, long‐term potentiation, inflammatory mediator regulation of transient receptor potential channels, and amphetamine addiction pathways. The protein–protein interaction analysis revealed four hub genes involved in the above pathways, including G protein subunit gamma 10, 5‐hydroxytryptamine receptor 2C, prodynorphin, and calcium/calmodulin‐dependent protein kinase II alpha. By combining multiscale correlation analyses and functional annotations, this study advances our understanding of the genetic and neural substrates of human neuroticism and emphasizes the importance of striatal functional properties in human neuroticism.

Permutation inference for canonical correlation analysis

Canonical correlation analysis (CCA) has become a key tool for population neuroimaging, allowing investigation of associations between many imaging and non-imaging measurements. As age, sex and other variables are often a source of variability not of direct interest, previous work has used CCA on residuals from a model that removes these effects, then proceeded directly to permutation inference. We show that a simple permutation test, as typically used to identify significant modes of shared variation on such data adjusted for nuisance variables, produces inflated error rates. The reason is that residualisation introduces dependencies among the observations that violate the exchangeability assumption. Even in the absence of nuisance variables, however, a simple permutation test for CCA also leads to excess error rates for all canonical correlations other than the first. The reason is that a simple permutation scheme does not ignore the variability already explained by previous canonical variables. Here we propose solutions for both problems: in the case of nuisance variables, we show that transforming the residuals to a lower dimensional basis where exchangeability holds results in a valid permutation test; for more general cases, with or without nuisance variables, we propose estimating the canonical correlations in a stepwise manner, removing at each iteration the variance already explained, while dealing with different number of variables in both sides. We also discuss how to address the multiplicity of tests, proposing an admissible test that is not conservative, and provide a complete algorithm for permutation inference for CCA.

Finding the needle in a high-dimensional haystack: Canonical correlation analysis for neuroscientists

The 21st century marks the emergence of "big data" with a rapid increase in the availability of datasets with multiple measurements. In neuroscience, brain-imaging datasets are more commonly accompanied by dozens or hundreds of phenotypic subject descriptors on the behavioral, neural, and genomic level. The complexity of such "big data" repositories offer new opportunities and pose new challenges for systems neuroscience. Canonical correlation analysis (CCA) is a prototypical family of methods that is useful in identifying the links between variable sets from different modalities. Importantly, CCA is well suited to describing relationships across multiple sets of data, such as in recently available big biomedical datasets. Our primer discusses the rationale, promises, and pitfalls of CCA.

Multiple Holdouts With Stability: Improving the Generalizability of Machine Learning Analyses of Brain-Behavior Relationships

BACKGROUND

In 2009, the National Institute of Mental Health launched the Research Domain Criteria, an attempt to move beyond diagnostic categories and ground psychiatry within neurobiological constructs that combine different levels of measures (e.g., brain imaging and behavior). Statistical methods that can integrate such multimodal data, however, are often vulnerable to overfitting, poor generalization, and difficulties in interpreting the results.

METHODS

We propose an innovative machine learning framework combining multiple holdouts and a stability criterion with regularized multivariate techniques, such as sparse partial least squares and kernel canonical correlation analysis, for identifying hidden dimensions of cross-modality relationships. To illustrate the approach, we investigated structural brain-behavior associations in an extensively phenotyped developmental sample of 345 participants (312 healthy and 33 with clinical depression). The brain data consisted of whole-brain voxel-based gray matter volumes, and the behavioral data included item-level self-report questionnaires and IQ and demographic measures.

RESULTS

Both sparse partial least squares and kernel canonical correlation analysis captured two hidden dimensions of brain-behavior relationships: one related to age and drinking and the other one related to depression. The applied machine learning framework indicates that these results are stable and generalize well to new data. Indeed, the identified brain-behavior associations are in agreement with previous findings in the literature concerning age, alcohol use, and depression-related changes in brain volume.

CONCLUSIONS

Multivariate techniques (such as sparse partial least squares and kernel canonical correlation analysis) embedded in our novel framework are promising tools to link behavior and/or symptoms to neurobiology and thus have great potential to contribute to a biologically grounded definition of psychiatric disorders.

Data-Driven Approaches to Neuroimaging Analysis to Enhance Psychiatric Diagnosis and Therapy

Combining advanced neuroimaging with novel computational methods in network science and machine learning has led to increasingly meaningful descriptions of structure and function in both the normal and the abnormal brain, thereby contributing significantly to our understanding of psychiatric disorders as circuit dysfunctions. Despite its marked potential for psychiatric care, this approach has not yet extended beyond the research setting to any clinically useful applications. Here we review current developments in the study of neuroimaging data using network models and machine learning methods, with a focus on their promise in offering a framework for clinical translation. We discuss 3 potential contributions of these methods to psychiatric care: 1) a better understanding of psychopathology beyond current diagnostic boundaries; 2) individualized prediction of treatment response and prognosis; and 3) formal theories to guide the development of novel interventions. Finally, we highlight current obstacles and sketch a forward-looking perspective of how the application of machine learning and network modeling methods should proceed to accelerate their potential transformation of clinically useful tools.

Identification of neurobehavioural symptom groups based on shared brain mechanisms

Most psychopathological disorders develop in adolescence. The biological basis for this development is poorly understood. To enhance diagnostic characterization and develop improved targeted interventions, it is critical to identify behavioural symptom groups that share neural substrates. We ran analyses to find relationships between behavioural symptoms and neuroimaging measures of brain structure and function in adolescence. We found two symptom groups, consisting of anxiety/depression and executive dysfunction symptoms, respectively, that correlated with distinct sets of brain regions and inter-regional connections, measured by structural and functional neuroimaging modalities. We found that the neural correlates of these symptom groups were present before behavioural symptoms had developed. These neural correlates showed case-control differences in corresponding psychiatric disorders, depression and attention deficit hyperactivity disorder in independent clinical samples. By characterizing behavioural symptom groups based on shared neural mechanisms, our results provide a framework for developing a classification system for psychiatric illness that is based on quantitative neurobehavioural measures.

Detection of Dominant Intra-prostatic Lesions in Patients With Prostate Cancer Using an Artificial Neural Network and MR Multi-modal Radiomics Analysis

Purpose: The aim of this study was to identify and rank discriminant radiomics features extracted from MR multi-modal images to construct an adaptive model for characterization of Dominant Intra-prostatic Lesions (DILs) from normal prostatic gland tissues (NT).

Methods and Materials: Two cohorts were retrospectively studied: Group A consisted of 98 patients and Group B 19 patients. Two image modalities were acquired using a 3.0T MR scanner: Axial T2 Weighted (T2W) and axial diffusion weighted (DW) imaging. A linear regression method was used to construct apparent diffusion coefficient (ADC) maps from DW images. DILs and the NT in the mirrored location were drawn on each modality. One hundred and sixty-eight radiomics features were extracted from DILs and NT. A Partial-Least-Squares-Correlation (PLSC) with one-way ANOVA along with bootstrapping ratio techniques were recruited to identify and rank the most discriminant latent variables. An artificial neural network (ANN) was constructed based on the optimal latent variable feature to classify the DILs and NTs. Nineteen patients were randomly chosen to test the contour variability effect on the radiomics analysis and the performance of the ANN. Finally, the trained ANN and a two dimension (2D) convolutional sampling method were combined and used to estimate DIL-NT probability map for two test cases.

Results: Among 168 radiomics-based latent variables, only the first four variables of each modality in the PLSC space were found to be significantly different between the DILs and NTs. Area Under Receiver Operating Characteristic (AUROC), Positive Predictive and Negative Predictive values (PPV and NPV) for the conventional method were 94%, 0.95, and 0.92, respectively. When the feature vector was randomly permuted 10,000 times, a very strong permutation-invariant efficiency (p < 0.0001) was achieved. The radiomic-based latent variables of the NTs and DILs showed no statistically significant differences (F_statistic < Fc = 4.11 with Confidence Level of 95% for all 8 variables) against contour variability. Dice coefficients between DIL-NT probability map and physician contours for the two test cases were 0.82 and 0.71, respectively.

Conclusion: This study demonstrates the high performance of combining radiomics information extracted from multimodal MR information such as T2WI and ADC maps, and adaptive models to detect DILs in patients with PCa.

A Metabolic Obesity Profile Is Associated With Decreased Gray Matter Volume in Cognitively Healthy Older Adults

Obesity is a risk factor for cognitive decline and gray matter volume loss in aging. Studies have shown that different metabolic factors, e.g., dysregulated glucose metabolism and systemic inflammation, might mediate this association. Yet, even though these risk factors tend to co-occur, they have mostly been investigated separately, making it difficult to establish their joint contribution to gray matter volume structure in aging. Here, we therefore aimed to determine a metabolic profile of obesity that takes into account different anthropometric and metabolic measures to explain differences in gray matter volume in aging. We included 748 elderly, cognitively healthy participants (age range: 60 - 79 years, BMI range: 17 - 42 kg/m2) of the LIFE-Adult Study. All participants had complete information on body mass index, waist-to-hip ratio, glycated hemoglobin, total blood cholesterol, high-density lipoprotein, interleukin-6, C-reactive protein, adiponectin and leptin. Voxelwise gray matter volume was extracted from T1-weighted images acquired on a 3T Siemens MRI scanner. We used partial least squares correlation to extract latent variables with maximal covariance between anthropometric, metabolic and gray matter volume and applied permutation/bootstrapping and cross-validation to test significance and reliability of the result. We further explored the association of the latent variables with cognitive performance. Permutation tests and cross-validation indicated that the first pair of latent variables was significant and reliable. The metabolic profile was driven by negative contributions from body mass index, waist-to-hip ratio, glycated hemoglobin, C-reactive protein and leptin and a positive contribution from adiponectin. It positively covaried with gray matter volume in temporal, frontal and occipital lobe as well as subcortical regions and cerebellum. This result shows that a metabolic profile characterized by high body fat, visceral adiposity and systemic inflammation is associated with reduced gray matter volume and potentially reduced executive function in older adults. We observed the highest contributions for body weight and fat mass, which indicates that factors underlying sustained energy imbalance, like sedentary lifestyle or intake of energy-dense food, might be important determinants of gray matter structure in aging.

An Integrated Indicator System and Evaluation Model for Regional Sustainable Development

Regional sustainable development has become a worldwide issue in recent years, but there is no single and universally agreed method of choosing indicators for sustainable development assessment. The subjective selection of indicators will affect the results of assessment. Each evaluation method has its own advantages and disadvantages, and the methods used to determine indicator weight also differ. Regional sustainable development is a complex system, which is difficult to evaluate objectively and scientifically using a single method. Therefore, a new integrated indicator system and evaluation model is constructed here to more accurately reflect regional sustainable development level. The indicator system and evaluation model were constructed using a case study of 17 cities in Shandong Province, China. The indicator system includes 4 subsystems, i.e., economy, society, resource, and environment. These indicators were selected through correlation analysis and discrimination analysis. A back propagation neural network was applied to evaluate the respective scores of the 4 subsystems. The comprehensive score for regional sustainable development was evaluated using the analytic hierarchy process with entropy correction. The results show that sustainable development levels in these 17 cities show a gradually decreasing trend from east to west and from coast to inland. Cities with an underdeveloped economy usually display poor levels of social development and serious environmental pollution. Through the improvement of indicator screening, evaluation model, and result correction, the error caused by a single evaluation method can be reduced significantly. This new methodology for indicator selection and comprehensive evaluation provides a new perspective for the assessment of regional sustainable development.

Unique Mapping of Structural and Functional Connectivity on Cognition

The unique mapping of structural brain connectivity (SC) and functional brain connectivity (FC) on cognition is currently not well understood. It is not clear whether cognition is mapped via a global connectome pattern or instead is underpinned by several sets of distributed connectivity patterns. Moreover, we also do not know whether the spatial distributions of SC and FC that underlie cognition are overlapping or distinct. Here, we study the relationship between SC and FC and an array of psychological tasks in 609 subjects (males, 269; females, 340) from the Human Connectome Project. We identified several sets of connections that each uniquely map onto cognitive function. We found a small number of distributed SCs and a larger set of corticocortical and corticosubcortical FCs that express this association. Importantly, the SC and FC each show unique and distinct patterns of variance across subjects as they relate to cognition. The results suggest that a complete understanding of connectome underpinnings of cognition calls for a combination of the two modalities.SIGNIFICANCE STATEMENT Structural connectivity (SC), the physical white-matter inter-regional pathways in the brain, and functional connectivity (FC), the temporal coactivations between the activity of the brain regions, have each been studied extensively. Little is known, however, about the distribution of variance in connections as they relate to cognition. Here, in a large sample of subjects (N = 609), we showed that two sets of brain-behavior patterns capture the correlations between SC and FC with a wide range of cognitive tasks, respectively. These brain-behavior patterns reveal distinct sets of connections within the SC and the FC network and provide new evidence that SC and FC each provide unique information for cognition.

Strategies for integrated analysis in imaging genetics studies

Imaging Genetics (IG) integrates neuroimaging and genomic data from the same individual, deepening our knowledge of the biological mechanisms behind neurodevelopmental domains and neurological disorders. Although the literature on IG has exponentially grown over the past years, the majority of studies have mainly analyzed associations between candidate brain regions and individual genetic variants. However, this strategy is not designed to deal with the complexity of neurobiological mechanisms underlying behavioral and neurodevelopmental domains. Moreover, larger sample sizes and increased multidimensionality of this type of data represents a challenge for standardizing modeling procedures in IG research. This review provides a systematic update of the methods and strategies currently used in IG studies, and serves as an analytical framework for researchers working in this field. To complement the functionalities of the Neuroconductor framework, we also describe existing R packages that implement these methodologies. In addition, we present an overview of how these methodological approaches are applied in integrating neuroimaging and genetic data.

A data-driven investigation of relationships between bipolar psychotic symptoms and schizophrenia genome-wide significant genetic loci

The etiologies of bipolar disorder (BD) and schizophrenia include a large number of common risk alleles, many of which are shared across the disorders. BD is clinically heterogeneous and it has been postulated that the pattern of symptoms is in part determined by the particular risk alleles carried, and in particular, that risk alleles also confer liability to schizophrenia influence psychotic symptoms in those with BD. To investigate links between psychotic symptoms in BD and schizophrenia risk alleles we employed a data-driven approach in a genotyped and deeply phenotyped sample of subjects with BD. We used sparse canonical correlation analysis (sCCA) (Witten, Tibshirani, & Hastie, ) to analyze 30 psychotic symptoms, assessed with the OPerational CRITeria checklist, and 82 independent genome-wide significant single nucleotide polymorphisms (SNPs) identified by the Schizophrenia Working group of the Psychiatric Genomics Consortium for which we had data in our BD sample (3,903 subjects). As a secondary analysis, we applied sCCA to larger groups of SNPs, and also to groups of symptoms defined according to a published factor analyses of schizophrenia. sCCA analysis based on individual psychotic symptoms revealed a significant association (p = .033), with the largest weights attributed to a variant on chromosome 3 (rs11411529), chr3:180594593, build 37) and delusions of influence, bizarre behavior and grandiose delusions. sCCA analysis using the same set of SNPs supported association with the same SNP and the group of symptoms defined "factor 3" (p = .012). A significant association was also observed to the "factor 3" phenotype group when we included a greater number of SNPs that were less stringently associated with schizophrenia; although other SNPs contributed to the significant multivariate association result, the greatest weight remained assigned to rs11411529. Our results suggest that the canonical correlation is a useful tool to explore phenotype-genotype relationships. To the best of our knowledge, this is the first study to apply this approach to complex, polygenic psychiatric traits. The sparse canonical correlation approach offers the potential to include a larger number of fine-grained systematic descriptors, and to include genetic markers associated with other disorders that are genetically correlated with BD.

Identification of associations between genotypes and longitudinal phenotypes via temporally-constrained group sparse canonical correlation analysis

Motivation

Neuroimaging genetics identifies the relationships between genetic variants (i.e., the single nucleotide polymorphisms) and brain imaging data to reveal the associations from genotypes to phenotypes. So far, most existing machine-learning approaches are widely used to detect the effective associations between genetic variants and brain imaging data at one time-point. However, those associations are based on static phenotypes and ignore the temporal dynamics of the phenotypical changes. The phenotypes across multiple time-points may exhibit temporal patterns that can be used to facilitate the understanding of the degenerative process. In this article, we propose a novel temporally constrained group sparse canonical correlation analysis (TGSCCA) framework to identify genetic associations with longitudinal phenotypic markers.

Results

The proposed TGSCCA method is able to capture the temporal changes in brain from longitudinal phenotypes by incorporating the fused penalty, which requires that the differences between two consecutive canonical weight vectors from adjacent time-points should be small. A new efficient optimization algorithm is designed to solve the objective function. Furthermore, we demonstrate the effectiveness of our algorithm on both synthetic and real data (i.e., the Alzheimer’s Disease Neuroimaging Initiative cohort, including progressive mild cognitive impairment, stable MCI and Normal Control participants). In comparison with conventional SCCA, our proposed method can achieve strong associations and discover phenotypic biomarkers across multiple time-points to guide disease-progressive interpretation.

Availability and implementation

The Matlab code is available at https://sourceforge.net/projects/ibrain-cn/files/.

Fast and Accurate Detection of Complex Imaging Genetics Associations Based on Greedy Projected Distance Correlation

Recent advances in imaging genetics produce large amounts of data including functional MRI images, single nucleotide polymorphisms (SNPs), and cognitive assessments. Understanding the complex interactions among these heterogeneous and complementary data has the potential to help with diagnosis and prevention of mental disorders. However, limited efforts have been made due to the high dimensionality, group structure, and mixed type of these data. In this paper we present a novel method to detect conditional associations between imaging genetics data. We use projected distance correlation to build a conditional dependency graph among high-dimensional mixed data, then use multiple testing to detect significant group level associations (e.g., ROI-gene). In addition, we introduce a scalable algorithm based on orthogonal greedy algorithm, yielding the greedy projected distance correlation (G-PDC). This can reduce the computational cost, which is critical for analyzing large-volume of imaging genomics data. The results from our simulations demonstrate a higher degree of accuracy with GPDC than distance correlation, Pearson's correlation and partial correlation, especially when the correlation is nonlinear. Finally, we apply our method to the Philadelphia Neurodevelopmental data cohort with 866 samples including fMRI images and SNP profiles. The results uncover several statistically significant and biologically interesting interactions, which are further validated with many existing studies. The Matlab code is available at https://sites.google.com/site/jianfang86/gPDC.

Pattern Discovery in Brain Imaging Genetics via SCCA Modeling with a Generic Non-convex Penalty

Brain imaging genetics intends to uncover associations between genetic markers and neuroimaging quantitative traits. Sparse canonical correlation analysis (SCCA) can discover bi-multivariate associations and select relevant features, and is becoming popular in imaging genetic studies. The L1-norm function is not only convex, but also singular at the origin, which is a necessary condition for sparsity. Thus most SCCA methods impose [Formula: see text]-norm onto the individual feature or the structure level of features to pursuit corresponding sparsity. However, the [Formula: see text]-norm penalty over-penalizes large coefficients and may incurs estimation bias. A number of non-convex penalties are proposed to reduce the estimation bias in regression tasks. But using them in SCCA remains largely unexplored. In this paper, we design a unified non-convex SCCA model, based on seven non-convex functions, for unbiased estimation and stable feature selection simultaneously. We also propose an efficient optimization algorithm. The proposed method obtains both higher correlation coefficients and better canonical loading patterns. Specifically, these SCCA methods with non-convex penalties discover a strong association between the APOE e4 rs429358 SNP and the hippocampus region of the brain. They both are Alzheimer's disease related biomarkers, indicating the potential and power of the non-convex methods in brain imaging genetics.

Modeling a linkage between blood transcriptional expression and activity in brain regions to infer the phenotype of schizophrenia patients

Hundreds of genetic loci participate to schizophrenia liability. It is also known that impaired cerebral connectivity is directly related to the cognitive and affective disturbances in schizophrenia. How genetic susceptibility and brain neural networks interact to specify a pathological phenotype in schizophrenia remains elusive. Imaging genetics, highlighting brain variations, has proven effective to establish links between vulnerability loci and associated clinical traits. As previous imaging genetics works in schizophrenia have essentially focused on structural DNA variants, these findings could be blurred by epigenetic mechanisms taking place during gene expression. We explored the meaningful links between genetic data from peripheral blood tissues on one hand, and regional brain reactivity to emotion task assayed by blood oxygen level-dependent functional magnetic resonance imaging on the other hand, in schizophrenia patients and matched healthy volunteers. We applied Sparse Generalized Canonical Correlation Analysis to identify joint signals between two blocks of variables: (i) the transcriptional expression of 33 candidate genes, and (ii) the blood oxygen level-dependent activity in 16 region of interest. Results suggested that peripheral transcriptional expression is related to brain imaging variations through a sequential pathway, ending with the schizophrenia phenotype. Generalization of such an approach to larger data sets should thus help in outlining the pathways involved in psychiatric illnesses such as schizophrenia.

IMAGING

SEARCHING FOR LINKS TO AID DIAGNOSIS: Researchers explore links between the expression of genes associated with schizophrenia in blood cells and variations in brain activity during emotion processing. El Chérif Ibrahim and Eric Fakra at Aix-Marseille Université, France, and colleagues have developed a method to relate the expression levels of 33 schizophrenia susceptibility genes in blood cells and functional magnetic resonance imaging (fMRI) data obtained as individuals carry out a task that triggers emotional responses. Although they found no significant differences in the expression of genes between the 26 patients with schizophrenia and 26 healthy controls they examined, variations in activity in the superior temporal gyrus were strongly linked to schizophrenia-associated gene expression and presence of disease. Similar analyses of larger data sets will shed further light on the relationship between peripheral molecular changes and disease-related behaviors and ultimately, aid the diagnosis of neuropsychiatric disease.

Age-related brain structural alterations as an intermediate phenotype of psychosis

BACKGROUND

There is only limited agreement with respect to location, directionality and functional implications of brain structural alterations observed in patients with schizophrenia. Additionally, their link to occurrence of psychotic symptoms remains unclear. A viable way of addressing these questions is to examine populations in an at-risk mental state (ARMS) before the transition to psychosis.

METHODS

We tested for structural brain alterations in individuals in an ARMS compared with healthy controls and patients with first-episode psychosis (FEP) using voxel-based morphometry and measures of cortical thickness. Furthermore, we evaluated if these alterations were modified by age and whether they were linked to the observed clinical symptoms.

RESULTS

Our sample included 59 individuals with ARMS, 26 healthy controls and 59 patients with FEP. We found increased grey matter volume and cortical thickness in individuals with ARMS and a similar pattern of structural alterations in patients with FEP. We further found stronger age-related reductions in grey matter volume and cortical thickness in both patients with FEP and individuals with ARMS, linking these alterations to observed clinical symptoms.

LIMITATIONS

The ARMS group comprised subgroups with heterogeneous levels of psychosis risk and medication status. Furthermore, the cross-sectional nature of our study and the reduced number of older patients limit conclusions with respect to observed interactions with age.

CONCLUSION

Our findings on consistent structural alterations in individuals with ARMS and patients with FEP and their link to clinical symptoms have major implications for understanding their time of occurrence and relevance to psychotic symptoms. Interactions with age found for these alterations may explain the heterogeneity of findings reported in the literature.

Joint sparse canonical correlation analysis for detecting differential imaging genetics modules

MOTIVATION

Imaging genetics combines brain imaging and genetic information to identify the relationships between genetic variants and brain activities. When the data samples belong to different classes (e.g. disease status), the relationships may exhibit class-specific patterns that can be used to facilitate the understanding of a disease. Conventional approaches often perform separate analysis on each class and report the differences, but ignore important shared patterns.

RESULTS

In this paper, we develop a multivariate method to analyze the differential dependency across multiple classes. We propose a joint sparse canonical correlation analysis method, which uses a generalized fused lasso penalty to jointly estimate multiple pairs of canonical vectors with both shared and class-specific patterns. Using a data fusion approach, the method is able to detect differentially correlated modules effectively and efficiently. The results from simulation studies demonstrate its higher accuracy in discovering both common and differential canonical correlations compared to conventional sparse CCA. Using a schizophrenia dataset with 92 cases and 116 controls including a single nucleotide polymorphism (SNP) array and functional magnetic resonance imaging data, the proposed method reveals a set of distinct SNP-voxel interaction modules for the schizophrenia patients, which are verified to be both statistically and biologically significant.

AVAILABILITY AND IMPLEMENTATION

The Matlab code is available at https://sites.google.com/site/jianfang86/JSCCA CONTACT: wyp@tulane.eduSupplementary information: Supplementary data are available at Bioinformatics online.

Selective association between cortical thickness and reference abilities in normal aging

A previous study of reference abilities and cortical thickness reported that association between reference abilities and cortical thickness summarized over large ROIs suppressed was suppressed after controlling for mean cortical thickness and global cognition. In this manuscript, we showed that preserving detailed spatial patterns of cortical thickness can identify reference-ability-specific association besides the association explained by global cognition and mean cortical thickness. We identified associations between cortical thickness and 3 cognitive reference abilities after controlling for mean thickness, global cognition, and linear chronological age: (1) memory, (2) perceptual speed, and (3) vocabulary. Global cognition was correlated with mean overall thickness but also was found to have a regionally specific pattern of associations. Nonlinear associations between cortical thickness and cognition were not observed, neither were nonlinear age effects. Age-by-thickness interactions were also absent. This implies that all thickness-cognition relations and age associations are independent of age and that consequently no age range is inherently special, since brain-behavioral findings are invariant across the whole age range.

Random Projection for Fast and Efficient Multivariate Correlation Analysis of High-Dimensional Data: A New Approach

In recent years, the advent of great technological advances has produced a wealth of very high-dimensional data, and combining high-dimensional information from multiple sources is becoming increasingly important in an extending range of scientific disciplines. Partial Least Squares Correlation (PLSC) is a frequently used method for multivariate multimodal data integration. It is, however, computationally expensive in applications involving large numbers of variables, as required, for example, in genetic neuroimaging. To handle high-dimensional problems, dimension reduction might be implemented as pre-processing step. We propose a new approach that incorporates Random Projection (RP) for dimensionality reduction into PLSC to efficiently solve high-dimensional multimodal problems like genotype-phenotype associations. We name our new method PLSC-RP. Using simulated and experimental data sets containing whole genome SNP measures as genotypes and whole brain neuroimaging measures as phenotypes, we demonstrate that PLSC-RP is drastically faster than traditional PLSC while providing statistically equivalent results. We also provide evidence that dimensionality reduction using RP is data type independent. Therefore, PLSC-RP opens up a wide range of possible applications. It can be used for any integrative analysis that combines information from multiple sources.

Estimating multivariate similarity between neuroimaging datasets with sparse canonical correlation analysis: an application to perfusion imaging

An increasing number of neuroimaging studies are based on either combining more than one data modality (inter-modal) or combining more than one measurement from the same modality (intra-modal). To date, most intra-modal studies using multivariate statistics have focused on differences between datasets, for instance relying on classifiers to differentiate between effects in the data. However, to fully characterize these effects, multivariate methods able to measure similarities between datasets are needed. One classical technique for estimating the relationship between two datasets is canonical correlation analysis (CCA). However, in the context of high-dimensional data the application of CCA is extremely challenging. A recent extension of CCA, sparse CCA (SCCA), overcomes this limitation, by regularizing the model parameters while yielding a sparse solution. In this work, we modify SCCA with the aim of facilitating its application to high-dimensional neuroimaging data and finding meaningful multivariate image-to-image correspondences in intra-modal studies. In particular, we show how the optimal subset of variables can be estimated independently and we look at the information encoded in more than one set of SCCA transformations. We illustrate our framework using Arterial Spin Labeling data to investigate multivariate similarities between the effects of two antipsychotic drugs on cerebral blood flow.