The billions of neurons that make up the adult brain are organized into domains and circuits during development. High-resolution measurements such as those enabled by single-cell molecular profiling have revealed unexpected cellular diversity. Genomic tools are lending insight into mechanisms behind neurodevelopmental disorders. Briscoe and Marèn review the insights gained as big data analyses are applied to neurodevelopmental questions. Science , this issue p. [eaaz8627] ### BACKGROUND The formation of the nervous system represents an astonishing feat of self-organization that is compromised in neuropsychiatric conditions such as autism and schizophrenia. Despite impressive progress in neuroscience over the past decades, our understanding of how billions of neurons come together to form the nervous system and enable function and behavior is in its infancy, especially when it comes to the human brain. However, the field is at a turning point. The introduction of new technologies that produce large volumes of high-resolution measurements—big data—has the potential to revolutionize the study of brain development. ### ADVANCES A foundation of developmental neuroscience is the detailed and systematic description of the nervous system. New methods are documenting the cellular composition and organization of neural tissue with ever-increasing resolution. The development of high-throughput and automated microscopy methods is charting the connectivity of thousands of neurons, delineating the structure of whole regions of the nervous system. Technology is also emerging for the large-scale analysis of the activity of entire brain regions. But probably the most obvious impact of high-throughput techniques is in the development of single-cell molecular profiling. A variety of approaches are being used to produce genome-wide molecular surveys with single-cell resolution from adult and embryonic tissue. In particular, transcriptomic analyses of thousands to millions of cells are providing an unprecedented molecular characterization of the brain, revealing previously unrecognized cell types, allowing interspecies comparisons, and suggesting mechanisms that account for the developmental origin of the diversity and function of neural cell types. Most neuropsychiatric disorders have a prominent heritable component and arise from the altered developmental processes during the formation of the nervous system. Recent advances in human genetics are beginning to shed light on the genetic architecture of these disorders and suggest how genetic variation confers susceptibility to disease. Large-scale sequencing studies have revealed causes that range from large-effect heterozygous mutations to highly polygenic conditions. In addition, the contribution of de novo somatic mutations to neurodevelopmental diseases is being recognized. Nevertheless, progressing from genetic findings to underlying biological mechanisms has proved challenging, not least because in many cases identifying the cells relevant to a disease process has been difficult. In this context, a convergence between neurogenetics and developmental neurobiology, driven by the increased resolution of the molecular and genetic assays, is likely to improve our understanding of the origin of neurological disorders and provide insight into basic developmental mechanisms. Whereas new molecular and genomic tools contribute to the identification of plausible neurobiological mechanisms, methods based on the directed development of pluripotent stem cells offer experimental access to developing human neural tissue to test hypotheses. Rapid progress is being made in the development of techniques that produce specific neural cell types or more complex mixtures of cell types that mimic the development of specific regions of the central nervous system. Questions remain about the accuracy of these in vitro models, and validation and refinement continue. Notwithstanding this uncertainty, the potential to study the etiology of neurological disorders in human neural tissue is already providing important insights. ### OUTLOOK New perspectives are emerging on long-standing questions about the ontogeny, composition, and function of the nervous system. They are addressing fundamental conceptual questions, such as what constitutes a cell type, and revealing biological mechanisms responsible for neurological disorders. The comparison of nervous system development between multiple individuals could conceivably identify individual variation in, for example, neural connectivity patterns that underpin behavioral individuality and may enable the investigation of the complex biological mechanisms underlying individuality. Certainly, integrating data from anatomical, developmental, genetic, and molecular studies has the potential to link cellular processes to functional and behavioral consequences. This strategy would provide fundamental insight and offer a new vision to the field. ![Figure] Big data approaches in developmental neurobiology. New technologies that produce large volumes of high-resolution measurements are documenting gene expression, connectivity, and function in the developing brain with an unprecedented level of detail. In combination with large-scale genetic studies, big data approaches are transforming our ability to interrogate the developing brain and identify causal mechanisms for its associated disorders. The formation of the human brain, which contains nearly 100 billion neurons making an average of 1000 connections each, represents an astonishing feat of self-organization. Despite impressive progress, our understanding of how neurons form the nervous system and enable function is very fragmentary, especially for the human brain. New technologies that produce large volumes of high-resolution measurements—big data—are now being brought to bear on this problem. Single-cell molecular profiling methods allow the exploration of neural diversity with increasing spatial and temporal resolution. Advances in human genetics are shedding light on the genetic architecture of neurodevelopmental disorders, and new approaches are revealing plausible neurobiological mechanisms underlying these conditions. Here, we review the opportunities and challenges of integrating large-scale genomics and genetics for the study of brain development. : /lookup/doi/10.1126/science.aaz8627 : pending:yes
AI has embraced medical applications from its inception, and some of the earliest work in successful application of AI technology occurred in medical contexts. Medicine in the twenty-first century will be very different than medicine in the late twentieth century. Fortunately, the technical challenges to AI that emerge are similar, and the prospects for success are high.
Small-molecule screens aimed at identifying therapeutic candidates traditionally search for molecules that affect one to several outputs at most, limiting discovery of true disease-modifying drugs. Researchers developed a machine-learning approach to identify small molecules that broadly correct gene networks dysregulated in a human induced pluripotent stem cell disease model of a common form of heart disease involving the aortic valve. Gene network correction by the most efficacious therapeutic candidate generalized to primary aortic valve cells derived from more than 20 patients with sporadic aortic valve disease and prevented aortic valve disease in vivo in a mouse model. In high-throughput sequencing data, performance comparisons between computational tools are essential for making informed decisions at each step of a project. Simulations are a critical part of method comparisons, but for standard Illumina sequencing of genomic DNA, they are often oversimplified, which leads to optimistic results for most tools.
GO has been diagnostics and drug discovery. In this paper, we further extensively used to compute the similarity between genes our previous study on gene-disease relationship (details in section 3) [19, 20]. In this work, we use the specifically with the multifunctional genes. We investigate functional annotations of a gene from the Gene Ontology the multifunctional gene-disease relationship based on the Annotation (GOA) databases to compute the shortest published molecular function annotations of genes from distance (path length) between the Molecular Function the Gene Ontology which is the most comprehensive (mf) GO terms annotating the gene.
To analyze the vast genomic data -- totaling more than 300 terabytes -- the scientists used Shirokane, a supercomputer used specifically for life science research at the University of Tokyo. The study found that liver cancer is caused by mutations or abnormalities in nearly 40 genes, including more than 10 that had never before been linked to liver cancer.