Maternal smoking during pregnancy and cortical structure in children with attention-deficit/hyperactivity disorder.

Maternal smoking during pregnancy (MSDP) is considered a risk factor for ADHD. While the mechanisms underlying this association are not well understood, MSDP may impact the developing brain in ways that lead to ADHD. Here, we investigated the effect of prenatal smoking exposure on cortical brain structures in children with ADHD using two methods of assessing prenatal exposure: maternal recall and epigenetic typing. Exposure groups were defined according to: (1) maternal recall (+MSDP: n = 24; -MSDP: n = 85) and (2) epigenetic markers (EM) (+EM: n = 14 -EM: n = 21). CIVET-1.1.12 and RMINC were used to acquire cortical brain measurements and perform statistical analyses, respectively. The vertex with highest significance was tested for association with Continuous Performance Test (CPT) dimensions. While no differences of brain structures were identified between +MSDP and -MSDP, +EM children (n = 10) had significantly smaller surface area in the right orbitofrontal cortex (ROFc), middle temporal cortex (RTc) and parahippocampal gyrus (RPHg) (15% FDR) compared to -EM children (n = 20). Cortical surface area in the RPHg significantly correlated with CPT commission errors T-scores. This study suggests that molecular markers may better define exposure to environmental risks, as compared to human recall.

Correlation of the methylomic signature of smoking during pregnancy with clinical traits in ADHD.

BACKGROUND: Attention deficit/hyperactivity disorder (ADHD) is a highly prevalent childhood disorder. Maternal smoking during pregnancy is a replicated environmental risk factor for this disorder. It is also a robust modifier of gene methylation during the prenatal developmental period. In this study, we sought to identify loci differentially methylated by maternal smoking during pregnancy and relate their methylation levels to various behavioural and physical outcomes relevant to ADHD. METHODS: We extracted DNA from blood samples from children diagnosed with ADHD and deeply phenotyped. Genome-wide DNA methylation was assessed using Infinium MethylationEPIC BeadChip. Maternal smoking during pregnancy was self-declared and assessed retrospectively. RESULTS: Our sample included 231 children with ADHD. Statistically significant differences in DNA methylation between children exposed or not to maternal smoking during pregnancy were detected in 3457 CpGs. We kept 30 CpGs with at least 5% of methylation difference between the 2 groups for further analysis. Six genes were associated with varied phenotypes of clinical relevance to ADHD. The levels of DNA methylation in RUNX1 were positively correlated with the CBCL scores, and DNA methylation in MYO1G correlated positively with the score at the Conners rating scale. Methylation level in a CpG located in GFI1 correlated with birthweight, a risk factor for ADHD. Differentially methylated regions were also identified and confirmed the association of RUNX1 methylation levels with the CBCL score. LIMITATIONS: The study has several limitations, including the retrospective recall with self-report of maternal smoking during pregnancy as well as the grouping of individuals of varying age and developmental stage and of both males and females. In addition, the correlation design prevents the building of causation models. CONCLUSION: This study provides evidence for the association between the level of methylation at specific loci and quantitative dimensions highly relevant for ADHD as well as birth weight, a measure that has already been associated with increased risk for ADHD. Our results provide further support to public health educational initiatives to stop maternal smoking during pregnancy.

Covariance regression with random forests.

Capturing the conditional covariances or correlations among the elements of a multivariate response vector based on covariates is important to various fields including neuroscience, epidemiology and biomedicine. We propose a new method called Covariance Regression with Random Forests (CovRegRF) to estimate the covariance matrix of a multivariate response given a set of covariates, using a random forest framework. Random forest trees are built with a splitting rule specially designed to maximize the difference between the sample covariance matrix estimates of the child nodes. We also propose a significance test for the partial effect of a subset of covariates. We evaluate the performance of the proposed method and significance test through a simulation study which shows that the proposed method provides accurate covariance matrix estimates and that the Type-1 error is well controlled. An application of the proposed method to thyroid disease data is also presented. CovRegRF is implemented in a freely available R package on CRAN.

Inter-individual variability during neurodevelopment: an investigation of linear and nonlinear resting-state EEG features in an age-homogenous group of infants.

Electroencephalography measures are of interest in developmental neuroscience as potentially reliable clinical markers of brain function. Features extracted from electroencephalography are most often averaged across individuals in a population with a particular condition and compared statistically to the mean of a typically developing group, or a group with a different condition, to define whether a feature is representative of the populations as a whole. However, there can be large variability within a population, and electroencephalography features often change dramatically with age, making comparisons difficult. Combined with often low numbers of trials and low signal-to-noise ratios in pediatric populations, establishing biomarkers can be difficult in practice. One approach is to identify electroencephalography features that are less variable between individuals and are relatively stable in a healthy population during development. To identify such features in resting-state electroencephalography, which can be readily measured in many populations, we introduce an innovative application of statistical measures of variance for the analysis of resting-state electroencephalography data. Using these statistical measures, we quantified electroencephalography features commonly used to measure brain development-including power, connectivity, phase-amplitude coupling, entropy, and fractal dimension-according to their intersubject variability. Results from 51 6-month-old infants revealed that the complexity measures, including fractal dimension and entropy, followed by connectivity were the least variable features across participants. This stability was found to be greatest in the right parietotemporal region for both complexity feature, but no significant region of interest was found for connectivity feature. This study deepens our understanding of physiological patterns of electroencephalography data in developing brains, provides an example of how statistical measures can be used to analyze variability in resting-state electroencephalography in a homogeneous group of healthy infants, contributes to the establishment of robust electroencephalography biomarkers of neurodevelopment through the application of variance analyses, and reveals that nonlinear measures may be most relevant biomarkers of neurodevelopment.

Genetic Heterogeneity Shapes Brain Connectivity in Psychiatry.

BACKGROUND: Polygenicity and genetic heterogeneity pose great challenges for studying psychiatric conditions. Genetically informed approaches have been implemented in neuroimaging studies to address this issue. However, the effects on functional connectivity of rare and common genetic risks for psychiatric disorders are largely unknown. Our objectives were to estimate and compare the effect sizes on brain connectivity of psychiatric genomic risk factors with various levels of complexity: oligogenic copy number variants (CNVs), multigenic CNVs, and polygenic risk scores (PRSs) as well as idiopathic psychiatric conditions and traits. METHODS: Resting-state functional magnetic resonance imaging data were processed using the same pipeline across 9 datasets. Twenty-nine connectome-wide association studies were performed to characterize the effects of 15 CNVs (1003 carriers), 7 PRSs, 4 idiopathic psychiatric conditions (1022 individuals with autism, schizophrenia, bipolar conditions, or attention-deficit/hyperactivity disorder), and 2 traits (31,424 unaffected control subjects). RESULTS: Effect sizes on connectivity were largest for psychiatric CNVs (estimates: 0.2-0.65 z score), followed by psychiatric conditions (0.15-0.42), neuroticism and fluid intelligence (0.02-0.03), and PRSs (0.01-0.02). Effect sizes of CNVs on connectivity were correlated to their effects on cognition and risk for disease (r = 0.9, p = 5.93 × 10-6). However, effect sizes of CNVs adjusted for the number of genes significantly decreased from small oligogenic to large multigenic CNVs (r = -0.88, p = 8.78 × 10-6). PRSs had disproportionately low effect sizes on connectivity compared with CNVs conferring similar risk for disease. CONCLUSIONS: Heterogeneity and polygenicity affect our ability to detect brain connectivity alterations underlying psychiatric manifestations.

A process convolution model for crash count data on a network.

Crash data observed on a road network often exhibit spatial correlation due to unobserved effects with inherent spatial correlation following the structure of the road network. It is important to model this spatial correlation while accounting for the road network structure. In this study, we introduce the network process convolution (NPC) model. In this model, the spatial correlation among crash data is captured by a Gaussian Process (GP) approximated through a kernel convolution approach. The GP's covariance function is based on path distance computed between a limited set of knots and crash data points on the road network. The proposed model offers a straightforward approach for predicting crash frequency at unobserved locations where covariates are available, and for interpolating the GP values anywhere on the network. Inference procedure is performed following the Bayesian paradigm and is implemented in R-INLA, which offers an estimation procedure that is very efficient compared to Markov Chain Monte Carlo sampling algorithms. We fitted our model to synthetic data and to crash data from Ottawa, Canada. We compared the proposed approach with a proper Conditional Autoregressive (pCAR) model, and with Poisson Regression (PR) and Negative Binomial (NB) models without latent effects. The results of the study indicated that although the pCAR model has comparable fitting performance, the NPC model outperforms pCAR when the main goal is to predict unobserved locations of interest. The proposed model also offers lower mean absolute error rates for cross validated crash counts, latent variable values, fixed-effect coefficients, as well as shorter interval scores for singletons. The NPC provides a natural way to account for the road network structure when considering the inclusion of spatially structured latent random effects in the modelling of crash data. It also offers an improved predictive capability for crash data on a road network.

Do stop-signs improve the safety for all road users? A before-after study of stop-controlled intersections using video-based trajectories and surrogate measures of safety.

Converting minor-approach-only stop (MAS) intersections to all-way-stop (AWS) intersections is a prevailing safety countermeasure in North American urban areas. Although the general population positively perceives the installation of stop-signs in residential areas, little research has investigated the impact of AWS on road safety and road user behaviour. This paper investigated the safety effectiveness of converting MAS to AWS intersections using an observational before and after approach and surrogate measures of safety. More specifically, the safety impacts of AWS conversion were investigated using multiple indicators, including vehicle speed measures, vehicle-pedestrian, vehicle-cyclist, vehicles-vehicle interactions as well as yielding rates before and after the treatment implementation. A multi-level regression approach was adopted to determine the effect of stop signs controlling for built environments, traffic exposure, and intersection geometry factors as well as site-specific unobserved heterogeneity. A unique sample of 31 intersections were used in this before-after study. From this sample, video data were collected before and after implementing AWS. In total, 245 h of video were automatically processed and corrected using a specialized computer vision software. More than 68,000 (37,668 before and 31,305 after AWS treatment) road user trajectories were obtained from 104 approaches. The results show that the conversion of MAS to AWS intersections significantly decreased vehicle speed and increased post-encroachment time. This work also shows that implementing AWS significantly increased the yielding rates from 45.7% to 76.7% in MAS conditions and reduced the average speed of motor-vehicles. Using multi-level regression model, it is estimated that when the intersection was converted from MAS to AWS, the minimum speed in the major approaches was reduced by 60.0%.

Evaluating safety-influencing factors at stop-controlled intersections using automated video analysis.

INTRODUCTION: Although stop signs are popular in North America, they have become controversial in cities like Montreal, Canada where they are often installed to reduce vehicular speeds and improve pedestrian safety despite limited evidence demonstrating their effectiveness. The purpose of this study is to evaluate the impact of stop-control configuration (and other features) on safety using statistical models and surrogate measures of safety (SMoS), namely vehicle speed, time-to-collision (TTC), and post-encroachment time (PET), while controlling for features of traffic, geometry, and built environment. METHODS: This project leverages high-resolution user trajectories extracted from video data collected for 100 intersections, 336 approaches, and 130,000 road users in Montreal to develop linear mixed-effects regression models to account for within-site and within-approach correlations. This research proposes the Intersection Exposure Group (IEG) indicator, an original method for classifying microscopic exposure of pedestrians and vehicles. RESULTS: Stop signs were associated with an average decrease in approach speed of 17.2 km/h and 20.1 km/h, at partially and fully stop-controlled respectively. Cyclist or pedestrian presence also significantly lower vehicle speeds. The proposed IEG measure was shown to successfully distinguish various types of pedestrian-vehicle interactions, allowing for the effect of each interaction type to vary in the model. CONCLUSIONS: The presence of stop signs significantly reduced approach speeds compared to uncontrolled approaches. Though several covariates were significantly related to TTC and PET for vehicle pairs, the models were unable to demonstrate a significant relationship between stop signs and vehicle-pedestrian interactions. Therefore, drawing conclusions regarding pedestrian safety is difficult. Practical Applications: As pedestrian safety is frequently used to justify new stop sign installations, this result has important policy implications. Policies implementing stop signs to reduce pedestrian crashes may be less effective than other interventions. Enforcement and education efforts, along with geometric design considerations, should accompany any changes in traffic control.

Massive GNSS data for road safety analysis: Comparing crash models for several Canadian cities and data sources.

Mobile sensors are a useful data source with applications in several transportation fields. Though cost of collection, transmission, and storage has limited studies on driving data and safety, this can be overcome through usage-based insurance (UBI). In UBI programs, drivers are monitored, and their premiums are adjusted based on driver-level surrogate safety measures (SSMs) related to exposure and driving style. Contextual link-level SSMs (volume, speed, or density) could further improve discount calibration. This study quantifies relationships between contextual SSMs and crashes and includes the validation of previous results (correlations between SSMs and crashes and statistical models estimated using smartphone-collected data from Quebec City) and the comparison of three Canadian cities (using UBI data from Quebec City, Montreal, and Ottawa). Extracted SSMs were compared to large volumes of historical crash frequency data using Spearman's Rank Correlation Coefficient and then implemented into spatial Bayesian crash models. Results from the UBI data generally matched those from the previous study, with observed correlations mirroring previous results in direction (braking, congestion, and speed variation are positively associated with crash frequency while mean speed is negatively associated) while correlation strength was slightly higher. Furthermore, these results were consistent between cities. For the crash modelling, repeatability of previous results in Quebec City was moderately good for the UBI data. Importantly for large-scale implementation, models estimated using UBI data were largely consistent between cities. This work provides an important contribution to the existing literature, clearly demonstrating how contextual safety measures could be applied to benefit UBI practices.

Maturational trajectories of pericortical contrast in typical brain development.

In the last few years, a significant amount of work has aimed to characterize maturational trajectories of cortical development. The role of pericortical microstructure putatively characterized as the gray-white matter contrast (GWC) at the pericortical gray-white matter boundary and its relationship to more traditional morphological measures of cortical morphometry has emerged as a means to examine finer grained neuroanatomical underpinnings of cortical changes. In this work, we characterize the GWC developmental trajectories in a representative sample (n = 394) of children and adolescents (~4 to ~22 years of age), with repeated scans (1-3 scans per subject, total scans n = 819). We tested whether linear, quadratic, or cubic trajectories of contrast development best described changes in GWC. A best-fit model was identified vertex-wise across the whole cortex via the Akaike Information Criterion (AIC). GWC across nearly the whole brain was found to significantly change with age. Cubic trajectories were likeliest for 63% of vertices, quadratic trajectories were likeliest for 20% of vertices, and linear trajectories were likeliest for 16% of vertices. A main effect of sex was observed in some regions, where males had a higher GWC than females. However, no sex by age interactions were found on GWC. In summary, our results suggest a progressive decrease in GWC at the pericortical boundary throughout childhood and adolescence. This work contributes to efforts seeking to characterize typical, healthy brain development and, by extension, can help elucidate aberrant developmental trajectories.

Sex-dependent complex association of TPH2 with multiple dimensions of ADHD.

BACKGROUND: Tryptophan hydroxylase 2 (TPH2) is a key enzyme in the biosynthesis of serotonin in the brain. This study aims to investigate the role of a functional variant in TPH2 (rs17110747) in the pathophysiology of ADHD. This variant has been implicated in mood disorders in recent meta-analysis. This study uses a comprehensive approach that combines association testing and pharmaco-dynamic evaluation of behaviour, in a large sample of children with ADHD (n = 570). METHODS: The association between various ADHD relevant traits and rs17110747 was analyzed using family-based association tests (FBAT). Children were assessed by parents, teachers and research staff under three experimental conditions (EC): baseline, placebo, and methylphenidate using a double-blind placebo-controlled crossover trial. OUTCOMES: FBAT analysis conducted in a sample stratified based on sex of the proband, showed that there was a highly significant overtransmission of the G allele from parents to affected girls. In addition, significant association with several behavioral and cognitive dimensions of ADHD was observed only when the proband was female. Further, girls with the G/G genotype (rs17110747) had greater response to placebo when evaluated by parents. INTERPRETATION: These results suggest that there may be a complex association of TPH2 in the etiology of ADHD, with a sex-specific effect.

Conditional canonical correlation estimation based on covariates with random forests.

MOTIVATION: Investigating the relationships between two sets of variables helps to understand their interactions and can be done with canonical correlation analysis (CCA). However, the correlation between the two sets can sometimes depend on a third set of covariates, often subject-related ones such as age, gender or other clinical measures. In this case, applying CCA to the whole population is not optimal and methods to estimate conditional CCA, given the covariates, can be useful. RESULTS: We propose a new method called Random Forest with Canonical Correlation Analysis (RFCCA) to estimate the conditional canonical correlations between two sets of variables given subject-related covariates. The individual trees in the forest are built with a splitting rule specifically designed to partition the data to maximize the canonical correlation heterogeneity between child nodes. We also propose a significance test to detect the global effect of the covariates on the relationship between two sets of variables. The performance of the proposed method and the global significance test is evaluated through simulation studies that show it provides accurate canonical correlation estimations and well-controlled Type-1 error. We also show an application of the proposed method with EEG data. AVAILABILITY AND IMPLEMENTATION: RFCCA is implemented in a freely available R package on CRAN (https://CRAN.R-project.org/package=RFCCA). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Machine Learning-Based Individualized Survival Prediction Model for Total Knee Replacement in Osteoarthritis: Data From the Osteoarthritis Initiative.

OBJECTIVE: By using machine learning, our study aimed to build a model to predict risk and time to total knee replacement (TKR) of an osteoarthritic knee. METHODS: Features were from the Osteoarthritis Initiative (OAI) cohort at baseline. Using the lasso method for variable selection in the Cox regression model, we identified the 10 most important characteristics among 1,107 features. The prognostic power of the selected features was assessed by the Kaplan-Meier method and applied to 7 machine learning methods: Cox, DeepSurv, random forests algorithm, linear/kernel support vector machine (SVM), and linear/neural multi-task logistic regression models. As some of the 10 first-found features included similar radiographic measurements, we further looked at using the least number of features without compromising the accuracy of the model. Prediction performance was assessed by the concordance index, Brier score, and time-dependent area under the curve (AUC). RESULTS: Ten features were identified and included radiographs, bone marrow lesions of the medial condyle on magnetic resonance imaging, hyaluronic acid injection, performance measure, medical history, and knee-related symptoms. The methodologies Cox, DeepSurv, and linear SVM demonstrated the highest accuracy (concordance index scores of 0.85, Brier score of 0.02, and an AUC of 0.87). DeepSurv was chosen to build the prediction model to estimate the time to TKR for a given knee. Moreover, we were able to decrease the features to only 3 and maintain the high accuracy (concordance index of 0.85, Brier score of 0.02, and AUC of 0.86), which included bone marrow lesions, Kellgren/Lawrence grade, and knee-related symptoms, to predict risk and time of a TKR event. CONCLUSION: For the first time, we developed a model using the OAI cohort to predict with high accuracy if a given osteoarthritic knee would require TKR, when a TKR would be required, and who would likely progress fast toward this event.

Association between COMT methylation and response to treatment in children with ADHD.

BACKGROUND: COMT had been considered a promising candidate gene in pharmacogenetic studies in ADHD; yet the findings from these studies have been inconsistent. Part of these inconsistencies could be related to epigenetic mechanisms (including DNA methylation). Here we investigated the role of genetic variants of the COMT gene on the methylation levels of CpG sites in the same gene and explored the effect of methylation on methylphenidate (MPH) and placebo (PBO) response in children with ADHD. METHODS: Two hundred and thirty children with ADHD (6-12 years) participated in a randomized, double-blind, placebo-controlled crossover trial with MPH. Univariate analysis was performed to examine the associations between genotypes in the COMT gene and DNA methylation in the same genetic loci. Association between the DNA methylation of 11 CpG sites and PBO/MPH responses were then assessed using spearman's correlation analysis in 212 children. Multiple linear regression analyses were performed to test the interaction between these factors while accounting for sex. RESULTS: Associations were observed between specific genetic variants and methylation level of cg20709110. Homozygous genotypes of GG (rs6269), CC (rs4633), GG (rs4818), Val/Val (rs4680) and the haplotype (ACCVal/GCGVal) were significantly associated with higher level of methylation. This CpG showed a significant correlation with placebo response (r = -0.15, P = 0.045) according to the teachers' evaluation, and a close-to significance correlation with response to MPH according to parents' evaluation (r = -0.134, p = 0.051). Regression analysis showed that in the model including rs4818, sex and DNA methylation of cg20709110 contributed significantly to treatment response. CONCLUSIONS: These preliminary results could provide evidence for the effect of genetic variations on methylation level and the involvement of the epigenetic variation of COMT loci in modulating the response to treatment in ADHD. TRIAL REGISTRATION: clinicaltrials.gov, number NCT00483106.

Genome-wide analysis of gene dosage in 24,092 individuals estimates that 10,000 genes modulate cognitive ability.

Genomic copy number variants (CNVs) are routinely identified and reported back to patients with neuropsychiatric disorders, but their quantitative effects on essential traits such as cognitive ability are poorly documented. We have recently shown that the effect size of deletions on cognitive ability can be statistically predicted using measures of intolerance to haploinsufficiency. However, the effect sizes of duplications remain unknown. It is also unknown if the effect of multigenic CNVs are driven by a few genes intolerant to haploinsufficiency or distributed across tolerant genes as well. Here, we identified all CNVs > 50 kilobases in 24,092 individuals from unselected and autism cohorts with assessments of general intelligence. Statistical models used measures of intolerance to haploinsufficiency of genes included in CNVs to predict their effect size on intelligence. Intolerant genes decrease general intelligence by 0.8 and 2.6 points of intelligence quotient when duplicated or deleted, respectively. Effect sizes showed no heterogeneity across cohorts. Validation analyses demonstrated that models could predict CNV effect sizes with 78% accuracy. Data on the inheritance of 27,766 CNVs showed that deletions and duplications with the same effect size on intelligence occur de novo at the same frequency. We estimated that around 10,000 intolerant and tolerant genes negatively affect intelligence when deleted, and less than 2% have large effect sizes. Genes encompassed in CNVs were not enriched in any GOterms but gene regulation and brain expression were GOterms overrepresented in the intolerant subgroup. Such pervasive effects on cognition may be related to emergent properties of the genome not restricted to a limited number of biological pathways.

Effect Sizes of Deletions and Duplications on Autism Risk Across the Genome.

OBJECTIVE: Deleterious copy number variants (CNVs) are identified in up to 20% of individuals with autism. However, levels of autism risk conferred by most rare CNVs remain unknown. The authors recently developed statistical models to estimate the effect size on IQ of all CNVs, including undocumented ones. In this study, the authors extended this model to autism susceptibility. METHODS: The authors identified CNVs in two autism populations (Simons Simplex Collection and MSSNG) and two unselected populations (IMAGEN and Saguenay Youth Study). Statistical models were used to test nine quantitative variables associated with genes encompassed in CNVs to explain their effects on IQ, autism susceptibility, and behavioral domains. RESULTS: The "probability of being loss-of-function intolerant" (pLI) best explains the effect of CNVs on IQ and autism risk. Deleting 1 point of pLI decreases IQ by 2.6 points in autism and unselected populations. The effect of duplications on IQ is threefold smaller. Autism susceptibility increases when deleting or duplicating any point of pLI. This is true for individuals with high or low IQ and after removing de novo and known recurrent neuropsychiatric CNVs. When CNV effects on IQ are accounted for, autism susceptibility remains mostly unchanged for duplications but decreases for deletions. Model estimates for autism risk overlap with previously published observations. Deletions and duplications differentially affect social communication, behavior, and phonological memory, whereas both equally affect motor skills. CONCLUSIONS: Autism risk conferred by duplications is less influenced by IQ compared with deletions. The model applied in this study, trained on CNVs encompassing >4,500 genes, suggests highly polygenic properties of gene dosage with respect to autism risk and IQ loss. These models will help to interpret CNVs identified in the clinic.

A novel statistical method for modeling covariate effects in bisulfite sequencing derived measures of DNA methylation.

Identifying disease-associated changes in DNA methylation can help us gain a better understanding of disease etiology. Bisulfite sequencing allows the generation of high-throughput methylation profiles at single-base resolution of DNA. However, optimally modeling and analyzing these sparse and discrete sequencing data is still very challenging due to variable read depth, missing data patterns, long-range correlations, data errors, and confounding from cell type mixtures. We propose a regression-based hierarchical model that allows covariate effects to vary smoothly along genomic positions and we have built a specialized EM algorithm, which explicitly allows for experimental errors and cell type mixtures, to make inference about smooth covariate effects in the model. Simulations show that the proposed method provides accurate estimates of covariate effects and captures the major underlying methylation patterns with excellent power. We also apply our method to analyze data from rheumatoid arthritis patients and controls. The method has been implemented in R package SOMNiBUS.

Stage-specific links between plasma neurofilament light and imaging biomarkers of Alzheimer's disease.

Neurofilament light (NfL) is a marker of neuroaxonal injury, a prominent feature of Alzheimer's disease. It remains uncertain, however, how it relates to amyloid and tau pathology or neurodegeneration across the Alzheimer's disease continuum. The aim of this study was to investigate how plasma NfL relates to amyloid and tau PET and MRI measures of brain atrophy in participants with and without cognitive impairment. We retrospectively examined the association between plasma NfL and MRI measures of grey/white matter volumes in the Alzheimer's Disease Neuroimaging Initiative [ADNI: n = 1149; 382 cognitively unimpaired control subjects and 767 cognitively impaired participants (mild cognitive impairment n = 420, Alzheimer's disease dementia n = 347)]. Longitudinal plasma NfL was measured using single molecule array (Simoa) technology. Cross-sectional associations between plasma NfL and PET amyloid and tau measures were independently assessed in two cohorts: ADNI [n = 198; 110 cognitively unimpaired, 88 cognitively impaired (MCI n = 67, Alzheimer's disease dementia n = 21), data accessed October 2018]; and Translational Biomarkers in Aging and Dementia [TRIAD, n = 116; 74 cognitively unimpaired, 42 cognitively impaired (MCI n = 16, Alzheimer's disease dementia n = 26), data obtained November 2017 to January 2019]. Associations between plasma NfL and imaging-derived measures were examined voxel-wise using linear regression (cross-sectional) and linear mixed effect models (longitudinal). Cross-sectional analyses in both cohorts showed that plasma NfL was associated with PET findings in brain regions typically affected by Alzheimer's disease; associations were specific to amyloid PET in cognitively unimpaired and tau PET in cognitively impaired (P < 0.05). Longitudinal analyses showed that NfL levels were associated with grey/white matter volume loss; grey matter atrophy in cognitively unimpaired was specific to APOE ε4 carriers (P < 0.05). These findings suggest that plasma NfL increases in response to amyloid-related neuronal injury in preclinical stages of Alzheimer's disease, but is related to tau-mediated neurodegeneration in symptomatic patients. As such, plasma NfL may a useful measure to monitor effects in disease-modifying drug trials.

Mutations associated with neuropsychiatric conditions delineate functional brain connectivity dimensions contributing to autism and schizophrenia.

16p11.2 and 22q11.2 Copy Number Variants (CNVs) confer high risk for Autism Spectrum Disorder (ASD), schizophrenia (SZ), and Attention-Deficit-Hyperactivity-Disorder (ADHD), but their impact on functional connectivity (FC) remains unclear. Here we report an analysis of resting-state FC using magnetic resonance imaging data from 101 CNV carriers, 755 individuals with idiopathic ASD, SZ, or ADHD and 1,072 controls. We characterize CNV FC-signatures and use them to identify dimensions contributing to complex idiopathic conditions. CNVs have large mirror effects on FC at the global and regional level. Thalamus, somatomotor, and posterior insula regions play a critical role in dysconnectivity shared across deletions, duplications, idiopathic ASD, SZ but not ADHD. Individuals with higher similarity to deletion FC-signatures exhibit worse cognitive and behavioral symptoms. Deletion similarities identified at the connectivity level could be related to the redundant associations observed genome-wide between gene expression spatial patterns and FC-signatures. Results may explain why many CNVs affect a similar range of neuropsychiatric symptoms.

Identification of the most important features of knee osteoarthritis structural progressors using machine learning methods.

OBJECTIVES: The aim was to identify the most important features of structural knee osteoarthritis (OA) progressors and classification using machine learning methods. METHODS: Participants, features and outcomes were from the Osteoarthritis Initiative. Features were from baseline (1107), including articular knee tissues (135) assessed by quantitative magnetic resonance imaging (MRI). OA progressors were ascertained by four outcomes: cartilage volume loss in medial plateau at 48 and 96 months (Prop_CV_48M, 96M), Kellgren-Lawrence (KL) grade ⩾ 2 and medial joint space narrowing (JSN) ⩾ 1 at 48 months. Six feature selection models were used to identify the common features in each outcome. Six classification methods were applied to measure the accuracy of the selected features in classifying the subjects into progressors and non-progressors. Classification of the best features was done using an automatic machine learning interface and the area under the curve (AUC). To prioritize the top five features, sparse partial least square (sPLS) method was used. RESULTS: For the classification of the best common features in each outcome, Multi-Layer Perceptron (MLP) achieved the highest AUC in Prop_CV_96M, KL and JSN (0.80, 0.88, 0.95), and Gradient Boosting Machine for Prop_CV_48M (0.70). sPLS showed the baseline top five features to predict knee OA progressors are the joint space width, mean cartilage thickness of the medial tibial plateau and sub-regions and JSN. CONCLUSION: In this comprehensive study using a large number of features (n = 1107) and MRI outcomes in addition to radiological outcomes, we identified the best features and classification methods for knee OA structural progressors. Data revealed baseline X-ray and MRI-based features could predict early OA knee progressors and that MLP is the best classification method.