Voxel-wise multivariate analysis of brain-psychosocial associations in adolescents reveals six latent dimensions of cognition and psychopathology.

BACKGROUND: Adolescence heralds the onset of much psychopathology, which may be conceptualized as an emergence of altered covariation between symptoms and brain measures. Multivariate methods can detect such modes of covariation or latent dimensions, but none specifically relating to psychopathology have yet been found using population-level structural brain data. Using voxel-wise (instead of parcellated) brain data may strengthen latent dimensions' brain-psychosocial relationships, but this creates computational challenges. METHODS: We obtained voxel-wise grey matter density and psychosocial variables from the baseline (aged 9-10 years) Adolescent Brain and Cognitive Development cohort (n=11288), and employed a state-of-the-art segmentation method, sparse partial least squares, and a rigorous machine learning framework to prevent overfitting. RESULTS: We found six latent dimensions, four pertaining specifically to mental health. The mental health dimensions related to overeating, anorexia/internalizing, oppositional symptoms (all p<0.002) and ADHD symptoms (p=0.03). ADHD related to increased and internalizing related to decreased grey matter density in dopaminergic and serotonergic midbrain areas, whereas oppositional symptoms related to increased grey matter in a noradrenergic nucleus. Internalizing related to increased and oppositional symptoms to reduced grey matter density in insula, cingulate and auditory cortices. Striatal regions featured strongly, with reduced caudate nucleus grey matter in ADHD, and reduced putamen grey matter in oppositional/conduct problems. Voxel-wise grey matter density generated stronger brain-psychosocial correlations than brain parcellations. CONCLUSIONS: Voxel-wise brain data strengthen latent dimensions of brain-psychosocial covariation and sparse multivariate methods increase their psychopathological specificity. Internalizing and externalizing are associated with opposite grey matter changes in similar cortical and subcortical areas.

Connectome-based reservoir computing with the conn2res toolbox.

The connection patterns of neural circuits form a complex network. How signaling in these circuits manifests as complex cognition and adaptive behaviour remains the central question in neuroscience. Concomitant advances in connectomics and artificial intelligence open fundamentally new opportunities to understand how connection patterns shape computational capacity in biological brain networks. Reservoir computing is a versatile paradigm that uses high-dimensional, nonlinear dynamical systems to perform computations and approximate cognitive functions. Here we present conn2res: an open-source Python toolbox for implementing biological neural networks as artificial neural networks. conn2res is modular, allowing arbitrary network architecture and dynamics to be imposed. The toolbox allows researchers to input connectomes reconstructed using multiple techniques, from tract tracing to noninvasive diffusion imaging, and to impose multiple dynamical systems, from spiking neurons to memristive dynamics. The versatility of the conn2res toolbox allows us to ask new questions at the confluence of neuroscience and artificial intelligence. By reconceptualizing function as computation, conn2res sets the stage for a more mechanistic understanding of structure-function relationships in brain networks.

A cross-cohort replicable and heritable latent dimension linking behaviour to multi-featured brain structure.

Identifying associations between interindividual variability in brain structure and behaviour requires large cohorts, multivariate methods, out-of-sample validation and, ideally, out-of-cohort replication. Moreover, the influence of nature vs nurture on brain-behaviour associations should be analysed. We analysed associations between brain structure (grey matter volume, cortical thickness, and surface area) and behaviour (spanning cognition, emotion, and alertness) using regularized canonical correlation analysis and a machine learning framework that tests the generalisability and stability of such associations. The replicability of brain-behaviour associations was assessed in two large, independent cohorts. The load of genetic factors on these associations was analysed with heritability and genetic correlation. We found one heritable and replicable latent dimension linking cognitive-control/executive-functions and positive affect to brain structural variability in areas typically associated with higher cognitive functions, and with areas typically associated with sensorimotor functions. These results revealed a major axis of interindividual behavioural variability linking to a whole-brain structural pattern.

Canonical Correlation Analysis and Partial Least Squares for Identifying Brain-Behavior Associations: A Tutorial and a Comparative Study.

Canonical correlation analysis (CCA) and partial least squares (PLS) are powerful multivariate methods for capturing associations across 2 modalities of data (e.g., brain and behavior). However, when the sample size is similar to or smaller than the number of variables in the data, standard CCA and PLS models may overfit, i.e., find spurious associations that generalize poorly to new data. Dimensionality reduction and regularized extensions of CCA and PLS have been proposed to address this problem, yet most studies using these approaches have some limitations. This work gives a theoretical and practical introduction into the most common CCA/PLS models and their regularized variants. We examine the limitations of standard CCA and PLS when the sample size is similar to or smaller than the number of variables. We discuss how dimensionality reduction and regularization techniques address this problem and explain their main advantages and disadvantages. We highlight crucial aspects of the CCA/PLS analysis framework, including optimizing the hyperparameters of the model and testing the identified associations for statistical significance. We apply the described CCA/PLS models to simulated data and real data from the Human Connectome Project and Alzheimer's Disease Neuroimaging Initiative (both of n > 500). We use both low- and high-dimensionality versions of these data (i.e., ratios between sample size and variables in the range of ∼1-10 and ∼0.1-0.01, respectively) to demonstrate the impact of data dimensionality on the models. Finally, we summarize the key lessons of the tutorial.

Audiovisual adaptation is expressed in spatial and decisional codes.

The brain adapts dynamically to the changing sensory statistics of its environment. Recent research has started to delineate the neural circuitries and representations that support this cross-sensory plasticity. Combining psychophysics and model-based representational fMRI and EEG we characterized how the adult human brain adapts to misaligned audiovisual signals. We show that audiovisual adaptation is associated with changes in regional BOLD-responses and fine-scale activity patterns in a widespread network from Heschl's gyrus to dorsolateral prefrontal cortices. Audiovisual recalibration relies on distinct spatial and decisional codes that are expressed with opposite gradients and time courses across the auditory processing hierarchy. Early activity patterns in auditory cortices encode sounds in a continuous space that flexibly adapts to misaligned visual inputs. Later activity patterns in frontoparietal cortices code decisional uncertainty consistent with these spatial transformations. Our findings suggest that regions within the auditory processing hierarchy multiplex spatial and decisional codes to adapt flexibly to the changing sensory statistics in the environment.

A hierarchical Bayesian model to find brain-behaviour associations in incomplete data sets.

Canonical Correlation Analysis (CCA) and its regularised versions have been widely used in the neuroimaging community to uncover multivariate associations between two data modalities (e.g., brain imaging and behaviour). However, these methods have inherent limitations: (1) statistical inferences about the associations are often not robust; (2) the associations within each data modality are not modelled; (3) missing values need to be imputed or removed. Group Factor Analysis (GFA) is a hierarchical model that addresses the first two limitations by providing Bayesian inference and modelling modality-specific associations. Here, we propose an extension of GFA that handles missing data, and highlight that GFA can be used as a predictive model. We applied GFA to synthetic and real data consisting of brain connectivity and non-imaging measures from the Human Connectome Project (HCP). In synthetic data, GFA uncovered the underlying shared and specific factors and predicted correctly the non-observed data modalities in complete and incomplete data sets. In the HCP data, we identified four relevant shared factors, capturing associations between mood, alcohol and drug use, cognition, demographics and psychopathological measures and the default mode, frontoparietal control, dorsal and ventral networks and insula, as well as two factors describing associations within brain connectivity. In addition, GFA predicted a set of non-imaging measures from brain connectivity. These findings were consistent in complete and incomplete data sets, and replicated previous findings in the literature. GFA is a promising tool that can be used to uncover associations between and within multiple data modalities in benchmark datasets (such as, HCP), and easily extended to more complex models to solve more challenging tasks.

Causal Inference in Audiovisual Perception.

In our natural environment the senses are continuously flooded with a myriad of signals. To form a coherent representation of the world, the brain needs to integrate sensory signals arising from a common cause and segregate signals coming from separate causes. An unresolved question is how the brain solves this binding or causal inference problem and determines the causal structure of the sensory signals. In this functional magnetic resonance imaging (fMRI) study human observers (female and male) were presented with synchronous auditory and visual signals at the same location (i.e., common cause) or different locations (i.e., separate causes). On each trial, observers decided whether signals come from common or separate sources(i.e., "causal decisions"). To dissociate participants' causal inference from the spatial correspondence cues we adjusted the audiovisual disparity of the signals individually for each participant to threshold accuracy. Multivariate fMRI pattern analysis revealed the lateral prefrontal cortex as the only region that encodes predominantly the outcome of observers' causal inference (i.e., common vs separate causes). By contrast, the frontal eye field (FEF) and the intraparietal sulcus (IPS0-4) form a circuitry that concurrently encodes spatial (auditory and visual stimulus locations), decisional (causal inference), and motor response dimensions. These results suggest that the lateral prefrontal cortex plays a key role in inferring and making explicit decisions about the causal structure that generates sensory signals in our environment. By contrast, informed by observers' inferred causal structure, the FEF-IPS circuitry integrates auditory and visual spatial signals into representations that guide motor responses.SIGNIFICANCE STATEMENT In our natural environment, our senses are continuously flooded with a myriad of signals. Transforming this barrage of sensory signals into a coherent percept of the world relies inherently on solving the causal inference problem, deciding whether sensory signals arise from a common cause and should hence be integrated or else be segregated. This functional magnetic resonance imaging study shows that the lateral prefrontal cortex plays a key role in inferring the causal structure of the environment. Crucially, informed by the spatial correspondence cues and the inferred causal structure the frontal eye field and the intraparietal sulcus form a circuitry that integrates auditory and visual spatial signals into representations that guide motor responses.

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.

Brain-behaviour modes of covariation in healthy and clinically depressed young people.

Understanding how variations in dimensions of psychometrics, IQ and demographics relate to changes in brain connectivity during the critical developmental period of adolescence and early adulthood is a major challenge. This has particular relevance for mental health disorders where a failure to understand these links might hinder the development of better diagnostic approaches and therapeutics. Here, we investigated this question in 306 adolescents and young adults (14-24 y, 25 clinically depressed) using a multivariate statistical framework, based on canonical correlation analysis (CCA). By linking individual functional brain connectivity profiles to self-report questionnaires, IQ and demographic data we identified two distinct modes of covariation. The first mode mapped onto an externalization/internalization axis and showed a strong association with sex. The second mode mapped onto a well-being/distress axis independent of sex. Interestingly, both modes showed an association with age. Crucially, the changes in functional brain connectivity associated with changes in these phenotypes showed marked developmental effects. The findings point to a role for the default mode, frontoparietal and limbic networks in psychopathology and depression.

PACAP application improves functional outcome of chronic retinal ischemic injury in rats-evidence from electroretinographic measurements.

Retinoprotective effects of pituitary adenylate cyclase activating polypeptide (PACAP) are well-known and have been demonstrated in various pathological conditions, such as diabetic retinopathy, excitotoxic retinal injury, UV light-induced degeneration, and ischemic retinal lesion. The neuronal degeneration observed in the different retinal layers under the above pathological conditions can be successfully decreased by PACAP; however, whether this morphological improvement is also reflected in functional amelioration remains unknown. Therefore, our purpose was to investigate the protective effect of PACAP on the rat retina after bilateral common carotid artery occlusion (BCCAO) with electroretinography (ERG) to parallel the functional data with the previous morphological and neurochemical observations. Control eyes received saline treatment while PACAP was injected into the vitreous space of the other eye immediately after the induction of ischemia. Retinal damage and protective effects of PACAP were quantified by the changes in the wave forms and amplitudes. On postoperative days 2 and 14, several parameters were assessed with special attention to the changes of b wave. The results confirm that the previously described morphological protection induced by PACAP treatment is reflected in functional improvement in ischemic retinal lesions.

Heat shock partially dissociates the overlapping modules of the yeast protein-protein interaction network: a systems level model of adaptation.

Network analysis became a powerful tool giving new insights to the understanding of cellular behavior. Heat shock, the archetype of stress responses, is a well-characterized and simple model of cellular dynamics. S. cerevisiae is an appropriate model organism, since both its protein-protein interaction network (interactome) and stress response at the gene expression level have been well characterized. However, the analysis of the reorganization of the yeast interactome during stress has not been investigated yet. We calculated the changes of the interaction-weights of the yeast interactome from the changes of mRNA expression levels upon heat shock. The major finding of our study is that heat shock induced a significant decrease in both the overlaps and connections of yeast interactome modules. In agreement with this the weighted diameter of the yeast interactome had a 4.9-fold increase in heat shock. Several key proteins of the heat shock response became centers of heat shock-induced local communities, as well as bridges providing a residual connection of modules after heat shock. The observed changes resemble to a 'stratus-cumulus' type transition of the interactome structure, since the unstressed yeast interactome had a globally connected organization, similar to that of stratus clouds, whereas the heat shocked interactome had a multifocal organization, similar to that of cumulus clouds. Our results showed that heat shock induces a partial disintegration of the global organization of the yeast interactome. This change may be rather general occurring in many types of stresses. Moreover, other complex systems, such as single proteins, social networks and ecosystems may also decrease their inter-modular links, thus develop more compact modules, and display a partial disintegration of their global structure in the initial phase of crisis. Thus, our work may provide a model of a general, system-level adaptation mechanism to environmental changes.

Network-based tools for the identification of novel drug targets.

In the past few years, network-based tools have become increasingly important in the identification of novel molecular targets for drug development. Systems-based approaches to predict signal transduction-related drug targets have developed into an especially promising field. Here, we summarize our studies, which indicate that modular bridges and overlaps of protein-protein interaction and signaling networks may be of key importance in future drug design. Intermodular nodes are very efficient in mediating the transmission of perturbations between signaling modules and are important in network cooperation. The analysis of stress-induced rearrangements of the yeast interactome by the ModuLand modularization algorithm indicated that components of modular overlap are key players in cellular adaptation to stress. Signaling crosstalk was much more pronounced in humans than in Caenorhabditis elegans or Drosophila melanogaster in the SignaLink (http://www.SignaLink.org) database, a uniformly curated database of eight major signaling pathways. We also showed that signaling proteins that participate in multiple pathways included multiple established drug targets and drug target candidates. Lastly, we caution that the pervasive overlap of cellular network modules implies that wider use of multitarget drugs to partially inhibit multiple individual proteins will be necessary to modify specific cellular functions, because targeting single proteins for complete disruption usually affects multiple cellular functions with little specificity for a particular process. Tools for analyzing network topology and especially network dynamics have great potential to identify alternative sets of targets for developing multitarget drugs.

Ageing as a price of cooperation and complexity: self-organization of complex systems causes the gradual deterioration of constituent networks.

The network concept is increasingly used for the description of complex systems. Here, we summarize key aspects of the evolvability and robustness of the hierarchical network set of macromolecules, cells, organisms and ecosystems. Listing the costs and benefits of cooperation as a necessary behaviour to build this network hierarchy, we outline the major hypothesis of the paper: the emergence of hierarchical complexity needs cooperation leading to the ageing (i.e. gradual deterioration) of the constituent networks. A stable environment develops cooperation leading to over-optimization, and forming an 'always-old' network, which accumulates damage, and dies in an apoptosis-like process. A rapidly changing environment develops competition forming a 'forever-young' network, which may suffer an occasional over-perturbation exhausting system resources, and causing death in a necrosis-like process. Giving a number of examples we demonstrate how cooperation evokes the gradual accumulation of damage typical to ageing. Finally, we show how various forms of cooperation and consequent ageing emerge as key elements in all major steps of evolution from the formation of protocells to the establishment of the globalized, modern human society.