Group Factor Analysis

Impact of digital screen media activity on functional brain organization in late childhood: Evidence from the ABCD study

The idea that the increased ubiquity of digital devices negatively impacts neurodevelopment is as compelling as it is disturbing. This study investigated this concern by systematically evaluating how different profiles of screen-based engagement related to functional brain organization in late childhood. We studied participants from a large and representative sample of young people participating in the first two years of the ABCD study (ages 9-12 years) to investigate the relations between self-reported use of various digital screen media activity (SMA) and functional brain organization. A series of generalized additive mixed models evaluated how these relationships related to functional outcomes associated with health and cognition. Of principal interest were two hypotheses: First, that functional brain organization (assessed through resting state functional connectivity MRI; rs-fcMRI) is related to digital screen engagement; and second, that children with higher rates of engagement will have functional brain organization profiles related to maladaptive functioning. Results did not support either of these predictions for SMA. Further, exploratory analyses predicting how screen media activity impacted neural trajectories showed no significant impact of SMA on neural maturation over a two-year period.

Zero-Shot Neural Decoding with Semi-Supervised Multi-View Embedding

Zero-shot neural decoding aims to decode image categories, which were not previously trained, from functional magnetic resonance imaging (fMRI) activity evoked when a person views images. However, having insufficient training data due to the difficulty in collecting fMRI data causes poor generalization capability. Thus, models suffer from the projection domain shift problem when novel target categories are decoded. In this paper, we propose a zero-shot neural decoding approach with semi-supervised multi-view embedding. We introduce the semi-supervised approach that utilizes additional images related to the target categories without fMRI activity patterns. Furthermore, we project fMRI activity patterns into a multi-view embedding space, i.e., visual and semantic feature spaces of viewed images to effectively exploit the complementary information. We define several source and target groups whose image categories are very different and verify the zero-shot neural decoding performance. The experimental results demonstrate that the proposed approach rectifies the projection domain shift problem and outperforms existing methods.

MC-RVAE: Multi-channel recurrent variational autoencoder for multimodal Alzheimer's disease progression modelling

The progression of neurodegenerative diseases, such as Alzheimer's Disease, is the result of complex mechanisms interacting across multiple spatial and temporal scales. Understanding and predicting the longitudinal course of the disease requires harnessing the variability across different data modalities and time, which is extremely challenging. In this paper, we propose a model based on recurrent variational autoencoders that is able to capture cross-channel interactions between different modalities and model temporal information. These are achieved thanks to its multi-channel architecture and its shared latent variational space, parametrized with a recurrent neural network. We evaluate our model on both synthetic and real longitudinal datasets, the latter including imaging and non-imaging data, with N=897 subjects. Results show that our multi-channel recurrent variational autoencoder outperforms a set of baselines (KNN, random forest, and group factor analysis) for the task of reconstructing missing modalities, reducing the mean absolute error by 5% (w.r.t. the best baseline) for both subcortical volumes and cortical thickness. Our model is robust to missing features within each modality and is able to generate realistic synthetic imaging biomarkers trajectories from cognitive scores.

RSC-based differential model with correlation removal for improving multi-omics clustering

Multi-omics clustering plays an important role in cancer subtyping. However, the data of different kinds of omics are often related, these correlations may reduce the clustering algorithm performance. It is crucial to eliminate the unexpected redundant information caused by these correlations between different omics. We proposed RSC-based differential model with correlation removal for improving multi-omics clustering (RSC-MCR). This method first introduced RSC to calculate the pairwise correlations of all features, and decomposed it to obtain the pairwise correlations of different omics features, thus built the connection between different omics based on the pairwise correlations of different omics features. Then, to remove the redundant correlation, we designed a differential model to calculate the degree of difference between the original feature matrix and the correlation matrix which contained the most relevant information between different omics. We compared the performance of RSC-MCR with decorrelation methods on different clustering methods (CC, FCM, SNF, NMF, LRAcluster). The experimental results on five cancer datasets show the efficiency of the RSC-MCR as well as improvements over other decorrelation methods.

Habituation or sensitization of brain response to food cues: Temporal dynamic analysis in an functional magnetic resonance imaging study

Introduction

In the modern obesogenic environment, heightened reactivity to food-associated cues plays a major role in overconsumption by evoking appetitive responses. Accordingly, functional magnetic resonance imaging (fMRI) studies have implicated regions of the salience and rewards processing in this dysfunctional food cue-reactivity, but the temporal dynamics of brain activation (sensitization or habituation over time) remain poorly understood.

Methods

Forty-nine obese or overweight adults were scanned in a single fMRI session to examine brain activation during the performance of a food cue-reactivity task. A general linear model (GLM) was used to validate the activation pattern of food cue reactivity in food > neutral contrast. The linear mixed effect models were used to examine the effect of time on the neuronal response during the paradigm of food cue reactivity. Neuro-behavioral relationships were investigated with Pearson's correlation tests and group factor analysis (GFA).

Results

A linear mixed-effect model revealed a trend for the time-by-condition interactions in the left medial amygdala [t(289) = 2.21, β = 0.1, P = 0.028], right lateral amygdala [t(289) = 2.01, β = 0.26, P = 0.045], right nucleus accumbens (NAc) [t(289) = 2.81, β = 0.13, P = 0.005] and left dorsolateral prefrontal cortex (DLPFC) [t(289) = 2.58, β = 0.14, P = 0.01], as well as in the left superior temporal cortex [42 Area: t(289) = 2.53, β = 0.15, P = 0.012; TE1.0_TE1.2 Area: t(289) = 3.13, β = 0.27, P = 0.002]. Habituation of blood-oxygenation-level-dependent (BOLD) signal during exposure to food vs. neutral stimuli was evident in these regions. We have not found any area in the brain with significant increased response to food-related cues over time (sensitization). Our results elucidate the temporal dynamics of cue-reactivity in overweight and obese individuals with food-induced craving. Both subcortical areas involved in reward processing and cortical areas involved in inhibitory processing are getting habituated over time in response to food vs. neutral cues. There were significant bivariate correlations between self-report behavioral/psychological measures with individual habituation slopes for the regions with dynamic activity, but no robust cross-unit latent factors were identified between the behavioral, demographic, and self-report psychological groups.

Discussion

This work provides novel insights into dynamic neural circuit mechanisms supporting food cue reactivity, thereby suggesting pathways in biomarker development and cue-desensitization interventions.

Applications of Artificial Intelligence to Obesity Research: Scoping Review of Methodologies

BACKGROUND

Obesity is a leading cause of preventable death worldwide. Artificial intelligence (AI), characterized by machine learning (ML) and deep learning (DL), has become an indispensable tool in obesity research.

OBJECTIVE

This scoping review aimed to provide researchers and practitioners with an overview of the AI applications to obesity research, familiarize them with popular ML and DL models, and facilitate the adoption of AI applications.

METHODS

We conducted a scoping review in PubMed and Web of Science on the applications of AI to measure, predict, and treat obesity. We summarized and categorized the AI methodologies used in the hope of identifying synergies, patterns, and trends to inform future investigations. We also provided a high-level, beginner-friendly introduction to the core methodologies to facilitate the dissemination and adoption of various AI techniques.

RESULTS

We identified 46 studies that used diverse ML and DL models to assess obesity-related outcomes. The studies found AI models helpful in detecting clinically meaningful patterns of obesity or relationships between specific covariates and weight outcomes. The majority (18/22, 82%) of the studies comparing AI models with conventional statistical approaches found that the AI models achieved higher prediction accuracy on test data. Some (5/46, 11%) of the studies comparing the performances of different AI models revealed mixed results, indicating the high contingency of model performance on the data set and task it was applied to. An accelerating trend of adopting state-of-the-art DL models over standard ML models was observed to address challenging computer vision and natural language processing tasks. We concisely introduced the popular ML and DL models and summarized their specific applications in the studies included in the review.

CONCLUSIONS

This study reviewed AI-related methodologies adopted in the obesity literature, particularly ML and DL models applied to tabular, image, and text data. The review also discussed emerging trends such as multimodal or multitask AI models, synthetic data generation, and human-in-the-loop that may witness increasing applications in obesity research.

[Development of Decomposition Approach for Comprehensive Understanding of Drug Effects]

As the term polypharmacology suggests, there are multiple actions of small-molecule compounds. We proposed a decomposition and understanding concept that sheds light on the small effects in comparison to the large effects by decomposing these multiple effects. This concept was embodied by describing the effects of the compounds in a transcriptome profile, followed by factor analysis to extract latent variables as decomposed effects. Application of this approach to public datasets resulted in the inferences of compound effects consistent with existing knowledge such as gene ontologies and pathways. In one experimental validation, the potential inducibility of endoplasmic reticulum stress of several commercial drugs was detected by decomposition. Another study successfully discriminated the effects of a natural product and its derivatives despite their structural similarity. In the era of big data, it is important to infer conceptual elements composed of measurable elements as a higher layer than the given data of a specimen, which can expand our perception and understanding of the specimen. This review introduces an example of such a philosophy by applying it to the multiple effects of drugs to contribute to the understanding.

Active coping strategies and less pre-pandemic alcohol use relate to college student mental health during the COVID-19 pandemic

Objective

To further delineate risk and resilience factors contributing to trajectories of mental health symptoms experienced by college students through the pandemic.

Participants

n = 183 college students (67.2% female).

Methods

Linear mixed models examined time effects on depression and anxiety. Propensity-matched subgroups exhibiting "increased" versus "low and stable" depression symptoms from before to after the pandemic-onset were compared on pre-pandemic demographic and psychological factors and COVID-related experiences and coping strategies.

Results

Students experienced worsening of mental health symptoms throughout the pandemic, particularly during Fall 2020 compared with Fall 2019 (Depression scale d = -0.43 [95% CI: -0.65 to -0.21]). The propensity-matched subgroup exhibiting relative resilience ("low and stable" symptoms) reported less alcohol use prior to the pandemic, greater use of active coping strategies, and less of an impact on their college progress.

Conclusions

Results point to several potential targets of screening and intervention to decrease residual impacts of the pandemic.

Variational Bayes for high-dimensional proportional hazards models with applications within gene expression

MOTIVATION

Few Bayesian methods for analyzing high-dimensional sparse survival data provide scalable variable selection, effect estimation and uncertainty quantification. Such methods often either sacrifice uncertainty quantification by computing maximum a posteriori estimates, or quantify the uncertainty at high (unscalable) computational expense.

RESULTS

We bridge this gap and develop an interpretable and scalable Bayesian proportional hazards model for prediction and variable selection, referred to as sparse variational Bayes. Our method, based on a mean-field variational approximation, overcomes the high computational cost of Markov chain Monte Carlo, whilst retaining useful features, providing a posterior distribution for the parameters and offering a natural mechanism for variable selection via posterior inclusion probabilities. The performance of our proposed method is assessed via extensive simulations and compared against other state-of-the-art Bayesian variable selection methods, demonstrating comparable or better performance. Finally, we demonstrate how the proposed method can be used for variable selection on two transcriptomic datasets with censored survival outcomes, and how the uncertainty quantification offered by our method can be used to provide an interpretable assessment of patient risk.

AVAILABILITY AND IMPLEMENTATION

our method has been implemented as a freely available R package survival.svb (https://github.com/mkomod/survival.svb).

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Task-Dependent Effective Connectivity of the Reward Network During Food Cue-Reactivity: A Dynamic Causal Modeling Investigation

Neural reactivity to food cues may play a central role in overeating and excess weight gain. Functional magnetic resonance imaging (fMRI) studies have implicated regions of the reward network in dysfunctional food cue-reactivity, but neural interactions underlying observed patterns of signal change remain poorly understood. Fifty overweight and obese participants with self-reported cue-induced food craving viewed food and neutral cues during fMRI scanning. Regions of the reward network with significantly greater food versus neutral cue-reactivity were used to specify plausible models of task-related neural interactions underlying the observed blood oxygenation level-dependent (BOLD) signal, and a bi-hemispheric winning model was identified in a dynamic causal modeling (DCM) framework. Neuro-behavioral correlations are investigated with group factor analysis (GFA) and Pearson’s correlation tests. The ventral tegmental area (VTA), amygdalae, and orbitofrontal cortices (OFC) showed significant food cue-reactivity. DCM suggests these activations are produced by largely reciprocal dynamic signaling between these regions, with food cues causing regional disinhibition and an apparent shifting of activity to the right amygdala. Intrinsic self-inhibition in the VTA and right amygdala is negatively correlated with measures of food craving and hunger and right-amygdalar disinhibition by food cues is associated with the intensity of cue-induced food craving, but no robust cross-unit latent factors were identified between the neural group and behavioral or demographic variable groups. Our results suggest a rich array of dynamic signals drive reward network cue-reactivity, with the amygdalae mediating much of the dynamic signaling between the VTA and OFCs. Neuro-behavioral correlations suggest particularly crucial roles for the VTA, right amygdala, and the right OFC-amygdala connection but the more robust GFA identified no cross-unit factors, so these correlations should be interpreted with caution. This investigation provides novel insights into dynamic circuit mechanisms with etiologic relevance to obesity, suggesting pathways in biomarker development and intervention.

Computational analyses of mechanism of action (MoA): data, methods and integration

The elucidation of a compound's Mechanism of Action (MoA) is a challenging task in the drug discovery process, but it is important in order to rationalise phenotypic findings and to anticipate potential side-effects. Bioinformatic approaches, advances in machine learning techniques and the increasing deposition of high-throughput data in public databases have significantly contributed to recent advances in the field, but it is not straightforward to decide which data and methods are most suitable to use in a given case. In this review, we focus on these methods and data and their applications in generating MoA hypotheses for subsequent experimental validation. We discuss compound-specific data such as -omics, cell morphology and bioactivity data, as well as commonly used supplementary prior knowledge such as network and pathway data, and provide information on databases where this data can be accessed. In terms of methodologies, we discuss both well-established methods (connectivity mapping, pathway enrichment) as well as more developing methods (neural networks and multi-omics integration). Finally, we review case studies where the MoA of a compound was successfully suggested from computational analysis by incorporating multiple data modalities and/or methodologies. Our aim for this review is to provide researchers with insights into the benefits and drawbacks of both the data and methods in terms of level of understanding, biases and interpretation - and to highlight future avenues of investigation which we foresee will improve the field of MoA elucidation, including greater public access to -omics data and methodologies which are capable of data integration.

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.

Understanding the Molecular Drivers of Disease Heterogeneity in Crohn's Disease Using Multi-omic Data Integration and Network Analysis

Crohn's disease (CD), a form of inflammatory bowel disease (IBD), is characterized by heterogeneity along multiple clinical axes, which in turn impacts disease progression and treatment modalities. Using advanced data integration approaches and systems biology tools, we studied the contribution of CD susceptibility variants and gene expression in distinct peripheral immune cell subsets (CD14+ monocytes and CD4+ T cells) to relevant clinical traits. Our analyses revealed that most clinical traits capturing CD heterogeneity could be associated with CD14+ and CD4+ gene expression rather than disease susceptibility variants. By disentangling the sources of variation, we identified molecular features that could potentially be driving the heterogeneity of various clinical traits of CD patients. Further downstream analyses identified contextual hub proteins such as genes encoding barrier functions, antimicrobial peptides, chemokines, and their receptors, which are either targeted by drugs used in CD or other inflammatory diseases or are relevant to the biological functions implicated in disease pathology. These hubs could be used as cell type-specific targets to treat specific subtypes of CD patients in a more individualized approach based on the underlying biology driving their disease subtypes. Our study highlights the importance of data integration and systems approaches to investigate complex and heterogeneous diseases such as IBD.

Computational principles and challenges in single-cell data integration

The development of single-cell multimodal assays provides a powerful tool for investigating multiple dimensions of cellular heterogeneity, enabling new insights into development, tissue homeostasis and disease. A key challenge in the analysis of single-cell multimodal data is to devise appropriate strategies for tying together data across different modalities. The term 'data integration' has been used to describe this task, encompassing a broad collection of approaches ranging from batch correction of individual omics datasets to association of chromatin accessibility and genetic variation with transcription. Although existing integration strategies exploit similar mathematical ideas, they typically have distinct goals and rely on different principles and assumptions. Consequently, new definitions and concepts are needed to contextualize existing methods and to enable development of new methods.

Failure to Identify Robust Latent Variables of Positive or Negative Valence Processing Across Units of Analysis

BACKGROUND

The heterogeneous nature of mood and anxiety disorders highlights a need for dimensionally based descriptions of psychopathology that inform better classification and treatment approaches. Following the Research Domain Criteria approach, this investigation sought to derive constructs assessing positive and negative valence domains across multiple units of analysis.

METHODS

Adults with clinically impairing mood and anxiety symptoms (N = 225) completed comprehensive assessments across several units of analysis. Self-report assessments included nine questionnaires that assess mood and anxiety symptoms and traits reflecting the negative and positive valence systems. Behavioral assessments included emotional reactivity and distress tolerance tasks, during which skin conductance and heart rate were measured. Neuroimaging assessments included fear conditioning and a reward processing task. The latent variable structure underlying these measures was explored using sparse Bayesian group factor analysis.

RESULTS

Group factor analysis identified 11 latent variables explaining 31.2% of the variance across tasks, none of which loaded across units of analysis or tasks. Instead, variance was best explained by individual latent variables for each unit of analysis within each task. Post hoc analyses 1) showed associations with small effect sizes between latent variables that were derived separately from functional magnetic resonance imaging and self-report data and 2) showed that some latent variables are not directly related to individual valence system constructs.

CONCLUSIONS

The lack of latent structure across units of analysis highlights challenges of the Research Domain Criteria approach and suggests that while dimensional analyses work well to reveal within-task features, more targeted approaches are needed to reveal latent cross-modal relationships that could illuminate psychopathology.

Financial Factors Determining the Investment Behavior of Lithuanian Business Companies

The article aims to identify and evaluate the financial factors influencing the investment behavior of Lithuanian companies. The article briefly reviews and summarizes previous research that provides detailed evidence of the financial factors that influence a firm’s investment behavior. The study is performed using correlation–regression and factor analysis. Sixteen Lithuanian joint-stock companies, the shares of which are listed on the Nasdaq Baltic stock exchange and whose main activity is not related to financial instruments, were selected for the research. Moreover, 58 companies are listed on the Nasdaq Baltic stock exchange (32 companies on the official list, 26 companies on the additional list). There are only 26 Lithuanian joint-stock companies in both lists. Out of 26 Lithuanian companies listed on this stock exchange, 16 were selected whose activities are not related to financial instruments. The results of the study provided strong evidence that a company’s financial assets have a positive impact on capital and overall profitability, i.e., Lithuanian companies with higher profitability invest in financial instruments more often, while companies with borrowed funds and with higher financial restrictions invest less. The study showed that the performance indicators of Lithuanian companies have a weak impact on the size of the company’s financial assets; therefore, it can be assumed that this is related to cognitive factors and heuristics.

Phenotype discovery from population brain imaging

•

A multimodal independent component analysis approach is presented for performing data fusion in UK biobank scale dataset.

•

This approach can estimate modes of population variability that enhance the ability to predict thousands of non-imaging phenotypes.

•

This approach improves predictive power compared with widely-used analysis strategies, single-modality decompositions and existing IDPs.

•

In UKB data, many interpretable associations with non-imaging phenotypes were identified.

Neuroimaging allows for the non-invasive study of the brain in rich detail. Data-driven discovery of patterns of population variability in the brain has the potential to be extremely valuable for early disease diagnosis and understanding the brain. The resulting patterns can be used as imaging-derived phenotypes (IDPs), and may complement existing expert-curated IDPs. However, population datasets, comprising many different structural and functional imaging modalities from thousands of subjects, provide a computational challenge not previously addressed. Here, for the first time, a multimodal independent component analysis approach is presented that is scalable for data fusion of voxel-level neuroimaging data in the full UK Biobank (UKB) dataset, that will soon reach 100,000 imaged subjects. This new computational approach can estimate modes of population variability that enhance the ability to predict thousands of phenotypic and behavioural variables using data from UKB and the Human Connectome Project. A high-dimensional decomposition achieved improved predictive power compared with widely-used analysis strategies, single-modality decompositions and existing IDPs. In UKB data (14,503 subjects with 47 different data modalities), many interpretable associations with non-imaging phenotypes were identified, including multimodal spatial maps related to fluid intelligence, handedness and disease, in some cases where IDP-based approaches failed.

Correspondence Between Perceived Pubertal Development and Hormone Levels in 9-10 Year-Olds From the Adolescent Brain Cognitive Development Study

Aim

To examine individual variability between perceived physical features and hormones of pubertal maturation in 9-10-year-old children as a function of sociodemographic characteristics.

Methods

Cross-sectional metrics of puberty were utilized from the baseline assessment of the Adolescent Brain Cognitive Development (ABCD) Study-a multi-site sample of 9-10 year-olds (n = 11,875)-and included perceived physical features via the pubertal development scale (PDS) and child salivary hormone levels (dehydroepiandrosterone and testosterone in all, and estradiol in females). Multi-level models examined the relationships among sociodemographic measures, physical features, and hormone levels. A group factor analysis (GFA) was implemented to extract latent variables of pubertal maturation that integrated both measures of perceived physical features and hormone levels.

Results

PDS summary scores indicated more males (70%) than females (31%) were prepubertal. Perceived physical features and hormone levels were significantly associated with child's weight status and income, such that more mature scores were observed among children that were overweight/obese or from households with low-income. Results from the GFA identified two latent factors that described individual differences in pubertal maturation among both females and males, with factor 1 driven by higher hormone levels, and factor 2 driven by perceived physical maturation. The correspondence between latent factor 1 scores (hormones) and latent factor 2 scores (perceived physical maturation) revealed synchronous and asynchronous relationships between hormones and concomitant physical features in this large young adolescent sample.

Conclusions

Sociodemographic measures were associated with both objective hormone and self-report physical measures of pubertal maturation in a large, diverse sample of 9-10 year-olds. The latent variables of pubertal maturation described a complex interplay between perceived physical changes and hormone levels that hallmark sexual maturation, which future studies can examine in relation to trajectories of brain maturation, risk/resilience to substance use, and other mental health outcomes.

Latent variables for region of interest activation during the monetary incentive delay task

BACKGROUND

The Monetary Incentive Delay task (MID) has been used extensively to probe anticipatory reward processes. However, individual differences evident during this task may relate to other constructs such as general arousal or valence processing (i.e., anticipation of negative versus positive outcomes). This investigation used a latent variable approach to parse activation patterns during the MID within a transdiagnostic clinical sample.

METHODS

Participants were drawn from the first 500 individuals recruited for the Tulsa-1000 (T1000), a naturalistic longitudinal study of 1000 participants aged 18-55 (n = 476 with MID data). We employed a multiview latent analysis method, group factor analysis, to characterize factors within and across variable sets consisting of: (1) region of interest (ROI)-based blood oxygenation level-dependent (BOLD) contrasts during reward and loss anticipation; and (2) self-report measures of positive and negative valence and related constructs.

RESULTS

Three factors comprised of ROI indicators emerged to accounted for >43% of variance and loaded on variables representing: (1) general arousal or general activation; (2) valence, with dissociable responses to anticipation of win versus loss; and (3) region-specific activation, with dissociable activation in salience versus perceptual brain networks. Two additional factors were comprised of self-report variables, which appeared to represent arousal and valence.

CONCLUSIONS

Results indicate that multiview techniques to identify latent variables offer a novel approach for differentiating brain activation patterns during task engagement. Such approaches may offer insight into neural processing patterns through dimension reduction, be useful for probing individual differences, and aid in the development of optimal explanatory or predictive frameworks.

The integration of social and neural synchrony: a case for ecologically valid research using MEG neuroimaging

The recent decade has seen a shift from artificial and environmentally deprived experiments in neuroscience to real-life studies on multiple brains in interaction, coordination and synchrony. In these new interpersonal synchrony experiments, there has been a growing trend to employ naturalistic social interactions to evaluate mechanisms underlying synchronous neuronal communication. Here, we emphasize the importance of integrating the assessment of neural synchrony with measurement of nonverbal behavioral synchrony as expressed in various social contexts: relaxed social interactions, planning a joint pleasurable activity, conflict discussion, invocation of trauma, or support giving and assess the integration of neural and behavioral synchrony across developmental stages and psychopathological conditions. We also showcase the advantages of magnetoencephalography neuroimaging as a promising tool for studying interactive neural synchrony and consider the challenge of ecological validity at the expense of experimental rigor. We review recent evidence of rhythmic information flow between brains in interaction and conclude with addressing state-of-the-art developments that may contribute to advance research on brain-to-brain coordination to the next level.

Brain anatomical covariation patterns linked to binge drinking and age at first full drink

•

We identified a reproducible cortical and subcortical brain structural covariation pattern.

•

A novel pattern discovery method Joint and Individual Variance Explained (JIVE) was used.

•

The cortical and subcortical structural covariation pattern is related to alcohol use initiation.

•

The identified pattern is dominated by covariation among brainstem, thalamus and PFC.

•

A thalamic-PFC-brainstem circuitry might be related to alcohol use initiation.

Binge drinking and age at first full drink (AFD) of alcohol prior to 21 years (AFD < 21) have been linked to neuroanatomical differences in cortical and subcortical grey matter (GM) volume, cortical thickness, and surface area. Despite the importance of understanding network-level relationships, structural covariation patterns among these morphological measures have yet to be examined in relation to binge drinking and AFD < 21. Here, we used the Joint and Individual Variance Explained (JIVE) method to characterize structural covariation patterns common across and specific to morphological measures in 293 participants (149 individuals with past-12-month binge drinking and 144 healthy controls) from the Human Connectome Project (HCP). An independent dataset (Nathan Kline Institute Rockland Sample; NKI-RS) was used to examine reproducibility/generalizability. We identified a reproducible joint component dominated by structural covariation between GM volume in the brainstem and thalamus proper, and GM volume and surface area in prefrontal cortical regions. Using linear mixed regression models, we found that participants with AFD < 21 showed lower joint component scores in both the HCP (beta = 0.059, p-value = 0.016; Cohen’s d = 0.441) and NKI-RS (beta = 0.023, p-value = 0.040, Cohen’s d = 0.216) datasets, whereas the individual thickness component associated with binge drinking (p-value = 0.02) and AFD < 21 (p-value < 0.001) in the HCP dataset was not statistically significant in the NKI-RS sample. Our findings were also generalizable to the HCP full sample (n = 880 participants). Taken together, our results show that use of JIVE analysis in high-dimensional, large-scale, psychiatry-related datasets led to discovery of a reproducible cortical and subcortical structural covariation pattern involving brain regions relevant to thalamic-PFC-brainstem neural circuitry which is related to AFD < 21 and suggests a possible extension of existing addiction neurocircuitry in humans.

Screen media activity does not displace other recreational activities among 9-10 year-old youth: a cross-sectional ABCD study®

Background

Screen media is among the most common recreational activities engaged in by children. The displacement hypothesis predicts that increased time spent on screen media activity (SMA) may be at the expense of engagement with other recreational activities, such as sport, music, and art. This study examined associations between non-educational SMA and recreational activity endorsement in 9–10-year-olds, when accounting for other individual (i.e., cognition, psychopathology), interpersonal (i.e., social environment), and sociodemographic characteristics.

Methods

Participants were 9254 youth from the Adolescent Brain Cognitive Development Study®. Latent factors reflecting SMA, cognition, psychopathology, and social environment were entered as independent variables into logistic mixed models. Sociodemographic covariates included age, sex, race/ethnicity, education, marital status, and household income. Outcome variables included any recreational activity endorsement (of 19 assessed), and specific sport (swimming, soccer, baseball) and hobby (music, art) endorsements.

Results

In unadjusted groupwise comparisons, youth who spent more time engaging with SMA were less likely to engage with other recreational activities (ps < .001). However, when variance in cognition, psychopathology, social environment, and sociodemographic covariates were accounted for, most forms of SMA were no longer significantly associated with recreational activity engagement (p > .05). Some marginal effects were observed: for every one SD increase in time spent on games and movies over more social forms of media, youth were at lower odds of engaging in recreational activities (adjusted odds ratio = 0·83, 95% CI 0·76–0·89). Likewise, greater general SMA was associated with lower odds of endorsing group-based sports, including soccer (0·93, 0·88–0·98) and baseball (0·92, 0·86–0·98). Model fit comparisons indicated that sociodemographic characteristics, particularly socio-economic status, explained more variance in rates of recreational activity engagement than SMA and other latent factors. Notably, youth from higher socio-economic families were up to 5·63 (3·83–8·29) times more likely to engage in recreational activities than youth from lower socio-economic backgrounds.

Conclusions

Results did not suggest that SMA largely displaces engagement in other recreational activities among 9–10-year-olds. Instead, socio-economic factors greatly contribute to rates of engagement. These findings are important considering recent shifts in time spent on SMA in childhood.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-020-09894-w.

Positive Economic, Psychosocial, and Physiological Ecologies Predict Brain Structure and Cognitive Performance in 9-10-Year-Old Children

While low socioeconomic status (SES) introduces risk for developmental outcomes among children, there are an array of proximal processes that determine the ecologies and thus the lived experiences of children. This study examined interrelations between 22 proximal measures in the economic, psychosocial, physiological, and perinatal ecologies of children, in association with brain structure and cognitive performance in a diverse sample of 8,158 9-10-year-old children from the Adolescent Brain Cognitive Development (ABCD) study. SES was measured by the income-to-needs ratio (INR), a measure used by federal poverty guidelines. Within the ABCD study, in what is one of the largest and most diverse cohorts of children studied in the United States, we replicate associations of low SES with lower total cortical surface area and worse cognitive performance. Associations between low SES (<200% INR) and measures of development showed the steepest increases with INR, with apparent increases still visible beyond the level of economic disadvantage in the range of 200-400% INR. Notably, we found three latent factors encompassing positive ecologies for children across the areas of economic, psychosocial, physiological, and perinatal well-being in association with better cognitive performance and the higher total cortical surface area beyond the effects of SES. Specifically, latent factors encompassing youth perceived social support and perinatal well-being were positive predictors of developmental measures for all children, regardless of SES. Further, we found a general latent factor that explained relationships between 20 of the proximal measures and encompassed a joint ecology of higher social and economic resources relative to low adversity across psychosocial, physiological, and perinatal domains. The association between the resource-to-adversity latent factor and cognitive performance was moderated by SES, such that for children in higher SES households, cognitive performance progressively increased with these latent factor scores, while for lower SES, cognitive performance increased only among children with the highest latent factor scores. Our findings suggest that both positive ecologies of increased access to resources and lower adversity are mutually critical for promoting better cognitive development in children from low SES households. Our findings inform future studies aiming to examine positive factors that influence healthier development in children.

An integrative machine learning approach to discovering multi-level molecular mechanisms of obesity using data from monozygotic twin pairs

We combined clinical, cytokine, genomic, methylation and dietary data from 43 young adult monozygotic twin pairs (aged 22-36 years, 53% female), where 25 of the twin pairs were substantially weight discordant (delta body mass index > 3 kg m^-2). These measurements were originally taken as part of the TwinFat study, a substudy of The Finnish Twin Cohort study. These five large multivariate datasets (comprising 42, 71, 1587, 1605 and 63 variables, respectively) were jointly analysed using an integrative machine learning method called group factor analysis (GFA) to offer new hypotheses into the multi-molecular-level interactions associated with the development of obesity. New potential links between cytokines and weight gain are identified, as well as associations between dietary, inflammatory and epigenetic factors. This encouraging case study aims to enthuse the research community to boldly attempt new machine learning approaches which have the potential to yield novel and unintuitive hypotheses. The source code of the GFA method is publically available as the R package GFA.

Covariate-driven factorization by thresholding for multiblock data

Multiblock data, where multiple groups of variables from different sources are observed for a common set of subjects, are routinely collected in many areas of science. Methods for joint factorization of such multiblock data are being developed to explore the potentially joint variation structure of the data. While most of the existing work focuses on delineating joint components, shared across all data blocks, from individual components, which is only relevant to a single data block, we propose to model and estimate partially joint components across some, but not all, data blocks. If covariates, with potential multiblock structures, are available, then the components are further modeled to be driven by the covariate information. To estimate such a covariate-driven, block-structured factor model, we propose an iterative algorithm based on thresholding, by transforming the problem of signal segmentation into a grouped variable selection problem. The proposed factorization provides accurate estimation of individual and (partially) joint structures in multiblock data, as confirmed by simulation studies. In the analysis of a real multiblock genomic dataset from the Cancer Genome Atlas project, we demonstrate that the estimated block structures provide straightforward interpretation and facilitate subsequent analyses.

LMSM: A modular approach for identifying lncRNA related miRNA sponge modules in breast cancer

Until now, existing methods for identifying lncRNA related miRNA sponge modules mainly rely on lncRNA related miRNA sponge interaction networks, which may not provide a full picture of miRNA sponging activities in biological conditions. Hence there is a strong need of new computational methods to identify lncRNA related miRNA sponge modules. In this work, we propose a framework, LMSM, to identify LncRNA related MiRNA Sponge Modules from heterogeneous data. To understand the miRNA sponging activities in biological conditions, LMSM uses gene expression data to evaluate the influence of the shared miRNAs on the clustered sponge lncRNAs and mRNAs. We have applied LMSM to the human breast cancer (BRCA) dataset from The Cancer Genome Atlas (TCGA). As a result, we have found that the majority of LMSM modules are significantly implicated in BRCA and most of them are BRCA subtype-specific. Most of the mediating miRNAs act as crosslinks across different LMSM modules, and all of LMSM modules are statistically significant. Multi-label classification analysis shows that the performance of LMSM modules is significantly higher than baseline’s performance, indicating the biological meanings of LMSM modules in classifying BRCA subtypes. The consistent results suggest that LMSM is robust in identifying lncRNA related miRNA sponge modules. Moreover, LMSM can be used to predict miRNA targets. Finally, LMSM outperforms a graph clustering-based strategy in identifying BRCA-related modules. Altogether, our study shows that LMSM is a promising method to investigate modular regulatory mechanism of sponge lncRNAs from heterogeneous data.

Previous studies have revealed that long non-coding RNAs (lncRNAs), as microRNA (miRNA) sponges or competing endogenous RNAs (ceRNAs), can regulate the expression levels of messenger RNAs (mRNAs) by decreasing the amount of miRNAs interacting with mRNAs. In this work, we hypothesize that the “tug-of-war” between RNA transcripts for attracting miRNAs is across groups or modules. Based on the hypothesis, we propose a framework called LMSM, to identify LncRNA related MiRNA Sponge Modules. Based on the two miRNA sponge modular competition principles, significant sharing of miRNAs and high canonical correlation between the sponge lncRNAs and mRNAs, LMSM is also capable of predicting miRNA targets. LMSM not only extends the ceRNA hypothesis, but also provides a novel way to investigate the biological functions and modular mechanism of lncRNAs in breast cancer.

Structural learning and integrative decomposition of multi-view data

The increased availability of multi-view data (data on the same samples from multiple sources) has led to strong interest in models based on low-rank matrix factorizations. These models represent each data view via shared and individual components, and have been successfully applied for exploratory dimension reduction, association analysis between the views, and consensus clustering. Despite these advances, there remain challenges in modeling partially-shared components and identifying the number of components of each type (shared/partially-shared/individual). We formulate a novel linked component model that directly incorporates partially-shared structures. We call this model SLIDE for Structural Learning and Integrative DEcomposition of multi-view data. The proposed model-fitting and selection techniques allow for joint identification of the number of components of each type, in contrast to existing sequential approaches. In our empirical studies, SLIDE demonstrates excellent performance in both signal estimation and component selection. We further illustrate the methodology on the breast cancer data from The Cancer Genome Atlas repository.

A review on machine learning principles for multi-view biological data integration

Driven by high-throughput sequencing techniques, modern genomic and clinical studies are in a strong need of integrative machine learning models for better use of vast volumes of heterogeneous information in the deep understanding of biological systems and the development of predictive models. How data from multiple sources (called multi-view data) are incorporated in a learning system is a key step for successful analysis. In this article, we provide a comprehensive review on omics and clinical data integration techniques, from a machine learning perspective, for various analyses such as prediction, clustering, dimension reduction and association. We shall show that Bayesian models are able to use prior information and model measurements with various distributions; tree-based methods can either build a tree with all features or collectively make a final decision based on trees learned from each view; kernel methods fuse the similarity matrices learned from individual views together for a final similarity matrix or learning model; network-based fusion methods are capable of inferring direct and indirect associations in a heterogeneous network; matrix factorization models have potential to learn interactions among features from different views; and a range of deep neural networks can be integrated in multi-modal learning for capturing the complex mechanism of biological systems.

Multiclass Nonnegative Matrix Factorization for Comprehensive Feature Pattern Discovery

In this big data era, interpretable machine learning models are strongly demanded for the comprehensive analytics of large-scale multiclass data. Characterizing all features from such data is a key but challenging step to understand the complexity. However, existing feature selection methods do not meet this need. In this paper, to address this problem, we propose a Bayesian multiclass nonnegative matrix factorization (MC-NMF) model with structured sparsity that is able to discover ubiquitous and class-specific features. Variational update rules were derived for efficient decomposition. In order to relieve the need of model selection and stably describe feature patterns, we further propose MC-NMF with stability selection, an ensemble method that collectively detects feature patterns from many runs of MC-NMF using different hyperparameter values and training subsets. We assessed our models on both simulated count data and multitumor ribonucleic acid-seq data. The experiments revealed that our models were able to recover predefined feature patterns from the simulated data and identify biologically meaningful patterns from the pan-cancer data.

Screen media activity and brain structure in youth: Evidence for diverse structural correlation networks from the ABCD study

The adolescent brain undergoes profound structural changes which is influenced by many factors. Screen media activity (SMA; e.g., watching television or videos, playing video games, or using social media) is a common recreational activity in children and adolescents; however, its effect on brain structure is not well understood. A multivariate approach with the first cross-sectional data release from the Adolescent Brain Cognitive Development (ABCD) study was used to test the maturational coupling hypothesis, i.e. the notion that coordinated patterns of structural change related to specific behaviors. Moreover, the utility of this approach was tested by determining the association between these structural correlation networks and psychopathology or cognition. ABCD participants with usable structural imaging and SMA data (N = 4277 of 4524) were subjected to a Group Factor Analysis (GFA) to identify latent variables that relate SMA to cortical thickness, sulcal depth, and gray matter volume. Subject scores from these latent variables were used in generalized linear mixed-effect models to investigate associations between SMA and internalizing and externalizing psychopathology, as well as fluid and crystalized intelligence. Four SMA-related GFAs explained 37% of the variance between SMA and structural brain indices. SMA-related GFAs correlated with brain areas that support homologous functions. Some but not all SMA-related factors corresponded with higher externalizing (Cohen's d effect size (ES) 0.06-0.1) but not internalizing psychopathology and lower crystalized (ES: 0.08-0.1) and fluid intelligence (ES: 0.04-0.09). Taken together, these findings support the notion of SMA related maturational coupling or structural correlation networks in the brain and provides evidence that individual differences of these networks have mixed consequences for psychopathology and cognitive performance.

Recovering Joint and Individual Components in Facial Data

A set of images depicting faces with different expressions or in various ages consists of components that are shared across all images (i.e., joint components) imparting to the depicted object the properties of human faces as well as individual components that are related to different expressions or age groups. Discovering the common (joint) and individual components in facial images is crucial for applications such as facial expression transfer and age progression. The problem is rather challenging when dealing with images captured in unconstrained conditions in the presence of sparse non-Gaussian errors of large magnitude (i.e., sparse gross errors or outliers) and contain missing data. In this paper, we investigate the use of a method recently introduced in statistics, the so-called Joint and Individual Variance Explained (JIVE) method, for the robust recovery of joint and individual components in visual facial data consisting of an arbitrary number of views. Since the JIVE is not robust to sparse gross errors, we propose alternatives, which are (1) robust to sparse gross, non-Gaussian noise, (2) able to automatically find the individual components rank, and (3) can handle missing data. We demonstrate the effectiveness of the proposed methods to several computer vision applications, namely facial expression synthesis and 2D and 3D face age progression 'in-the-wild'.

Generalized Bayesian Factor Analysis for Integrative Clustering with Applications to Multi-Omics Data

Integrative clustering is a clustering approach for multiple datasets, which provide different views of a common group of subjects. It enables analyzing multi-omics data jointly to, for example, identify the subtypes of diseases, cells, and so on, capturing the complex underlying biological processes more precisely. On the other hand, there has been a great deal of interest in incorporating the prior structural knowledge on the features into statistical analyses over the past decade. The knowledge on the gene regulatory network (pathways) can potentially be incorporated into many genomic studies. In this paper, we propose a novel integrative clustering method which can incorporate the prior graph knowledge. We first develop a generalized Bayesian factor analysis (GBFA) framework, a sparse Bayesian factor analysis which can take into account the graph information. Our GBFA framework employs the spike and slab lasso (SSL) prior to impose sparsity on the factor loadings and the Markov random field (MRF) prior to encourage smoothing over the adjacent factor loadings, which establishes a unified shrinkage adaptive to the loading size and the graph structure. Then, we use the framework to extend iCluster+, a factor analysis based integrative clustering approach. A novel variational EM algorithm is proposed to efficiently estimate the MAP estimator for the factor loadings. Extensive simulation studies and the application to the NCI60 cell line dataset demonstrate that the propose method is superior and delivers more biologically meaningful outcomes.

Multi-Omics Factor Analysis-a framework for unsupervised integration of multi-omics data sets

Multi-omics studies promise the improved characterization of biological processes across molecular layers. However, methods for the unsupervised integration of the resulting heterogeneous data sets are lacking. We present Multi-Omics Factor Analysis (MOFA), a computational method for discovering the principal sources of variation in multi-omics data sets. MOFA infers a set of (hidden) factors that capture biological and technical sources of variability. It disentangles axes of heterogeneity that are shared across multiple modalities and those specific to individual data modalities. The learnt factors enable a variety of downstream analyses, including identification of sample subgroups, data imputation and the detection of outlier samples. We applied MOFA to a cohort of 200 patient samples of chronic lymphocytic leukaemia, profiled for somatic mutations, RNA expression, DNA methylation and ex vivo drug responses. MOFA identified major dimensions of disease heterogeneity, including immunoglobulin heavy-chain variable region status, trisomy of chromosome 12 and previously underappreciated drivers, such as response to oxidative stress. In a second application, we used MOFA to analyse single-cell multi-omics data, identifying coordinated transcriptional and epigenetic changes along cell differentiation.

Patient similarity by joint matrix trifactorization to identify subgroups in acute myeloid leukemia

Objective

Computing patients’ similarity is of great interest in precision oncology since it supports clustering and subgroup identification, eventually leading to tailored therapies. The availability of large amounts of biomedical data, characterized by large feature sets and sparse content, motivates the development of new methods to compute patient similarities able to fuse heterogeneous data sources with the available knowledge.

Materials and Methods

In this work, we developed a data integration approach based on matrix trifactorization to compute patient similarities by integrating several sources of data and knowledge. We assess the accuracy of the proposed method: (1) on several synthetic data sets which similarity structures are affected by increasing levels of noise and data sparsity, and (2) on a real data set coming from an acute myeloid leukemia (AML) study. The results obtained are finally compared with the ones of traditional similarity calculation methods.

Results

In the analysis of the synthetic data set, where the ground truth is known, we measured the capability of reconstructing the correct clusters, while in the AML study we evaluated the Kaplan-Meier curves obtained with the different clusters and measured their statistical difference by means of the log-rank test. In presence of noise and sparse data, our data integration method outperform other techniques, both in the synthetic and in the AML data.

Discussion

In case of multiple heterogeneous data sources, a matrix trifactorization technique can successfully fuse all the information in a joint model. We demonstrated how this approach can be efficiently applied to discover meaningful patient similarities and therefore may be considered a reliable data driven strategy for the definition of new research hypothesis for precision oncology.

Conclusion

The better performance of the proposed approach presents an advantage over previous methods to provide accurate patient similarities supporting precision medicine.

Discriminative Bayesian Dictionary Learning for Classification

We propose a Bayesian approach to learn discriminative dictionaries for sparse representation of data. The proposed approach infers probability distributions over the atoms of a discriminative dictionary using a finite approximation of Beta Process. It also computes sets of Bernoulli distributions that associate class labels to the learned dictionary atoms. This association signifies the selection probabilities of the dictionary atoms in the expansion of class-specific data. Furthermore, the non-parametric character of the proposed approach allows it to infer the correct size of the dictionary. We exploit the aforementioned Bernoulli distributions in separately learning a linear classifier. The classifier uses the same hierarchical Bayesian model as the dictionary, which we present along the analytical inference solution for Gibbs sampling. For classification, a test instance is first sparsely encoded over the learned dictionary and the codes are fed to the classifier. We performed experiments for face and action recognition; and object and scene-category classification using five public datasets and compared the results with state-of-the-art discriminative sparse representation approaches. Experiments show that the proposed Bayesian approach consistently outperforms the existing approaches.

Sparse group factor analysis for biclustering of multiple data sources

MOTIVATION

Modelling methods that find structure in data are necessary with the current large volumes of genomic data, and there have been various efforts to find subsets of genes exhibiting consistent patterns over subsets of treatments. These biclustering techniques have focused on one data source, often gene expression data. We present a Bayesian approach for joint biclustering of multiple data sources, extending a recent method Group Factor Analysis to have a biclustering interpretation with additional sparsity assumptions. The resulting method enables data-driven detection of linear structure present in parts of the data sources.

RESULTS

Our simulation studies show that the proposed method reliably infers biclusters from heterogeneous data sources. We tested the method on data from the NCI-DREAM drug sensitivity prediction challenge, resulting in an excellent prediction accuracy. Moreover, the predictions are based on several biclusters which provide insight into the data sources, in this case on gene expression, DNA methylation, protein abundance, exome sequence, functional connectivity fingerprints and drug sensitivity.

AVAILABILITY AND IMPLEMENTATION

http://research.cs.aalto.fi/pml/software/GFAsparse/

CONTACTS

: kerstin.bunte@googlemail.com or samuel.kaski@aalto.fi.