Insights from an autism imaging biomarker challenge: Promises and threats to biomarker discovery.

MRI has been extensively used to identify anatomical and functional differences in Autism Spectrum Disorder (ASD). Yet, many of these findings have proven difficult to replicate because studies rely on small cohorts and are built on many complex, undisclosed, analytic choices. We conducted an international challenge to predict ASD diagnosis from MRI data, where we provided preprocessed anatomical and functional MRI data from > 2,000 individuals. Evaluation of the predictions was rigorously blinded. 146 challengers submitted prediction algorithms, which were evaluated at the end of the challenge using unseen data and an additional acquisition site. On the best algorithms, we studied the importance of MRI modalities, brain regions, and sample size. We found evidence that MRI could predict ASD diagnosis: the 10 best algorithms reliably predicted diagnosis with AUC∼0.80 - far superior to what can be currently obtained using genotyping data in cohorts 20-times larger. We observed that functional MRI was more important for prediction than anatomical MRI, and that increasing sample size steadily increased prediction accuracy, providing an efficient strategy to improve biomarkers. We also observed that despite a strong incentive to generalise to unseen data, model development on a given dataset faces the risk of overfitting: performing well in cross-validation on the data at hand, but not generalising. Finally, we were able to predict ASD diagnosis on an external sample added after the end of the challenge (EU-AIMS), although with a lower prediction accuracy (AUC=0.72). This indicates that despite being based on a large multisite cohort, our challenge still produced biomarkers fragile in the face of dataset shifts.

Polygenic Architecture of Human Neuroanatomical Diversity.

We analyzed the genomic architecture of neuroanatomical diversity using magnetic resonance imaging and single nucleotide polymorphism (SNP) data from >26 000 individuals from the UK Biobank project and 5 other projects that had previously participated in the ENIGMA (Enhancing NeuroImaging Genetics through Meta-Analysis) consortium. Our results confirm the polygenic architecture of neuroanatomical diversity, with SNPs capturing from 40% to 54% of regional brain volume variance. Chromosomal length correlated with the amount of phenotypic variance captured, r ~ 0.64 on average, suggesting that at a global scale causal variants are homogeneously distributed across the genome. At a local scale, SNPs within genes (~51%) captured ~1.5 times more genetic variance than the rest, and SNPs with low minor allele frequency (MAF) captured less variance than the rest: the 40% of SNPs with MAF <5% captured

Multitrait analysis to decipher the intertwined genetic architecture of neuroanatomical phenotypes and psychiatric disorders.

BACKGROUND: There is increasing evidence of shared genetic factors between psychiatric disorders and brain magnetic resonance imaging (MRI) phenotypes. However, deciphering the joint genetic architecture of these outcomes has proven challenging, and new approaches are needed to infer potential genetic structure underlying those phenotypes. Multivariate analyses is arising as a meaningful approach to reveal links between MRI phenotypes and psychiatric disorders missed by univariate approaches. METHODS: We first conducted univariate and multivariate genome-wide association studies (GWAS) for nine MRI-derived brain volume phenotypes in 20K UK Biobank participants. We next performed various complementary enrichment analyses to assess whether and how univariate and multitrait approaches can distinguish disorder-associated and non-disorder-associated variants from six psychiatric disorders: bipolarity, attention-deficit/hyperactivity disorder (ADHD), autism, schizophrenia, obsessive-compulsive disorder, and major depressive disorder. Finally, we conducted a clustering analysis of top associated variants based on their MRI multitrait association using an optimized k-medoids approach. RESULTS: Univariate MRI GWAS displayed only negligible genetic correlation with psychiatric disorders, while multitrait GWAS identified multiple new associations and showed significant enrichment for variants related to both ADHD and schizophrenia. Clustering analyses further detected two clusters displaying not only enrichment for association with ADHD and schizophrenia, but also consistent direction of effects. Functional annotation analyses of those clusters pointed to multiple potential mechanisms, suggesting in particular a role of neurotrophins pathways on both MRI and schizophrenia. CONCLUSIONS: Our results show that multitrait association signature can be used to infer genetically-driven latent MRI variables associated with psychiatric disorders, opening paths for future biomarker development.

Diversity and evolution of cerebellar folding in mammals.

The process of brain folding is thought to play an important role in the development and organisation of the cerebrum and the cerebellum. The study of cerebellar folding is challenging due to the small size and abundance of its folia. In consequence, little is known about its anatomical diversity and evolution. We constituted an open collection of histological data from 56 mammalian species and manually segmented the cerebrum and the cerebellum. We developed methods to measure the geometry of cerebellar folia and to estimate the thickness of the molecular layer. We used phylogenetic comparative methods to study the diversity and evolution of cerebellar folding and its relationship with the anatomy of the cerebrum. Our results show that the evolution of cerebellar and cerebral anatomy follows a stabilising selection process. We observed two groups of phenotypes changing concertedly through evolution: a group of 'diverse' phenotypes - varying over several orders of magnitude together with body size, and a group of 'stable' phenotypes varying over less than 1 order of magnitude across species. Our analyses confirmed the strong correlation between cerebral and cerebellar volumes across species, and showed in addition that large cerebella are disproportionately more folded than smaller ones. Compared with the extreme variations in cerebellar surface area, folial anatomy and molecular layer thickness varied only slightly, showing a much smaller increase in the larger cerebella. We discuss how these findings could provide new insights into the diversity and evolution of cerebellar folding, the mechanisms of cerebellar and cerebral folding, and their potential influence on the organisation of the brain across species.

BioConvert: a comprehensive format converter for life sciences.

Bioinformatics is a field known for the numerous standards and formats that have been developed over the years. This plethora of formats, sometimes complementary, and often redundant, poses many challenges to bioinformatics data analysts. They constantly need to find the best tool to convert their data into the suitable format, which is often a complex, technical and time consuming task. Moreover, these small yet important tasks are often difficult to make reproducible. To overcome these difficulties, we initiated BioConvert, a collaborative project to facilitate the conversion of life science data from one format to another. BioConvert aggregates existing software within a single framework and complemented them with original code when needed. It provides a common interface to make the user experience more streamlined instead of having to learn tens of them. Currently, BioConvert supports about 50 formats and 100 direct conversions in areas such as alignment, sequencing, phylogeny, and variant calling. In addition to being useful for end-users, BioConvert can also be utilized by developers as a universal benchmarking framework for evaluating and comparing numerous conversion tools. Additionally, we provide a web server implementing an online user-friendly interface to BioConvert, hence allowing direct use for the community.

Exploring the multidimensional nature of repetitive and restricted behaviors and interests (RRBI) in autism: neuroanatomical correlates and clinical implications.

BACKGROUND: Repetitive and restricted behaviors and interests (RRBI) are core symptoms of autism with a complex entity and are commonly categorized into 'motor-driven' and 'cognitively driven'. RRBI symptomatology depends on the individual's clinical environment limiting the understanding of RRBI physiology, particularly their associated neuroanatomical structures. The complex RRBI heterogeneity needs to explore the whole RRBI spectrum by integrating the clinical context [autistic individuals, their relatives and typical developing (TD) individuals]. We hypothesized that different RRBI dimensions would emerge by exploring the whole spectrum of RRBI and that these dimensions are associated with neuroanatomical signatures-involving cortical and subcortical areas. METHOD: A sample of 792 individuals composed of 267 autistic subjects, their 370 first-degree relatives and 155 TD individuals was enrolled in the study. We assessed the whole patterns of RRBI in each individual by using the Repetitive Behavior Scale-Revised and the Yale-Brown Obsessive Compulsive Scale. We estimated brain volumes using MRI scanner for a subsample of the subjects (n = 152, 42 ASD, 89 relatives and 13 TD). We first investigated the dimensionality of RRBI by performing a principal component analysis on all items of these scales and included all the sampling population. We then explored the relationship between RRBI-derived factors with brain volumes using linear regression models. RESULTS: We identified 3 main factors (with 30.3% of the RRBI cumulative variance): Factor 1 (FA1, 12.7%) reflected mainly the 'motor-driven' RRBI symptoms; Factor 2 and 3 (respectively, 8.8% and 7.9%) gathered mainly Y-BOCS related items and represented the 'cognitively driven' RRBI symptoms. These three factors were significantly associated with the right/left putamen volumes but with opposite effects: FA1 was negatively associated with an increased volume of the right/left putamen conversely to FA2 and FA3 (all uncorrected p < 0.05). FA1 was negatively associated with the left amygdala (uncorrected p < 0.05), and FA2 was positively associated with the left parietal structure (uncorrected p = 0.001). CONCLUSION: Our results suggested 3 coherent RRBI dimensions involving the putamen commonly and other structures according to the RRBI dimension. The exploration of the putamen's integrative role in RSBI needs to be strengthened in further studies.

Phenotypic effects of genetic variants associated with autism.

While over 100 genes have been associated with autism, little is known about the prevalence of variants affecting them in individuals without a diagnosis of autism. Nor do we fully appreciate the phenotypic diversity beyond the formal autism diagnosis. Based on data from more than 13,000 individuals with autism and 210,000 undiagnosed individuals, we estimated the odds ratios for autism associated to rare loss-of-function (LoF) variants in 185 genes associated with autism, alongside 2,492 genes displaying intolerance to LoF variants. In contrast to autism-centric approaches, we investigated the correlates of these variants in individuals without a diagnosis of autism. We show that these variants are associated with a small but significant decrease in fluid intelligence, qualification level and income and an increase in metrics related to material deprivation. These effects were larger for autism-associated genes than in other LoF-intolerant genes. Using brain imaging data from 21,040 individuals from the UK Biobank, we could not detect significant differences in the overall brain anatomy between LoF carriers and non-carriers. Our results highlight the importance of studying the effect of the genetic variants beyond categorical diagnosis and the need for more research to understand the association between these variants and sociodemographic factors, to best support individuals carrying these variants.

Morning Plasma Melatonin Differences in Autism: Beyond the Impact of Pineal Gland Volume.

While low plasma melatonin, a neuro-hormone synthesized in the pineal gland, has been frequently associated with autism, our understanding of the mechanisms behind it have remained unclear. In this exploratory study, we hypothesized that low melatonin levels in ASD could be linked to a decrease of the pineal gland volume (PGV). PGV estimates with magnetic resonance imaging (MRI) with a voxel-based volumetric measurement method and early morning plasma melatonin levels were evaluated for 215 participants, including 78 individuals with ASD, 90 unaffected relatives, and 47 controls. We first found that both early morning melatonin level and PGV were lower in patients compared to controls. We secondly built a linear model and observed that plasma melatonin was correlated to the group of the participant, but also to the PGV. To further understand the relationship between PGV and melatonin, we generated a normative model of the PGV relationship with melatonin level based on control participant data. We found an effect of PGV on normalized melatonin levels in ASD. Melatonin deficit appeared however more related to the group of the subject. Thus, melatonin variations in ASD could be mainly driven by melatonin pathway dysregulation.

Cerebellar Volume in Autism: Literature Meta-analysis and Analysis of the Autism Brain Imaging Data Exchange Cohort.

BACKGROUND: The neuroanatomical bases of autism spectrum disorder remain largely unknown. Among the most widely discussed candidate endophenotypes, differences in cerebellar volume have been often reported as statistically significant. METHODS: We aimed at objectifying this possible alteration by performing a systematic meta-analysis of the literature and an analysis of the ABIDE (Autism Brain Imaging Data Exchange) cohort. Our meta-analysis sought to determine a combined effect size of autism spectrum disorder diagnosis on different measures of the cerebellar anatomy as well as the effect of possible factors of variability across studies. We then analyzed the cerebellar volume of 328 patients and 353 control subjects from the ABIDE project. RESULTS: The meta-analysis of the literature suggested a weak but significant association between autism spectrum disorder diagnosis and increased cerebellar volume (p = .049, uncorrected), but the analysis of ABIDE did not show any relationship. The studies meta-analyzed were generally underpowered; however, the number of statistically significant findings was larger than expected. CONCLUSIONS: Although we could not provide a conclusive explanation for this excess of significant findings, our analyses would suggest publication bias as a possible reason. Finally, age, sex, and IQ were important sources of cerebellar volume variability, although independent of autism diagnosis.