Volume 21 Issue 9, September 2018

The surge in single-cell and single-nucleus RNA-sequencing has raised the question of the value of bulk tissue transcriptomics. Kelley et al. describe an analysis framework by which existing bulk transcriptomic data can be reanalyzed using cell-type-specific data to yield insights into cell-type variation across brain regions and diseases.

A study has found that young male mice with a mutation of the autism-associated gene Chd8 show abnormal behaviors and elevated neuronal activation in several brain areas under stressful conditions, while female mice with the same mutation have reduced baseline neuronal activity, which may protect them from developing these abnormal phenotypes.

A deep-learning-based software package called DeepLabCut rapidly and easily enables video-based motion tracking in any animal species. Such tracking technology is bound to revolutionize movement science and behavioral tracking in the laboratory and is also poised to find many applications in the real world.

The authors review recent work at the intersection of cognitive science, computational neuroscience and artificial intelligence that develops and tests computational models mimicking neural and cognitive function during a wide range of tasks.

A GWAS of lifetime cannabis use reveals new risk loci, shows that cannabis use has genetic overlap with smoking and alcohol use, and indicates that the likelihood of initiating cannabis use is causally influenced by schizophrenia.

The authors use integrative deconvolution of gene expression data to reveal core transcriptional features of CNS cell classes in humans, and identify cell-class-specific transcriptional differences in disease, among CNS regions, and between species.

The authors use single-nucleus RNA-seq to identify 10 GABAergic interneuron subtypes in human cortex layer 1. Molecular, morphological, and physiological evidence points to an emerging human cell type, the rosehip cell, not found in other species.

Phagocytes drive formation & resolution of neuroinflammatory lesions. The authors use in vivo imaging to track expression of proinflammatory & resolution enzymes & follow temporal changes in individual phagocytes polarization states in inflamed CNS.

The authors describe microscopic channels that directly connect the skull marrow cavities with the meninges. Neutrophils originating from the skull marrow have a higher propensity to travel to the ischemic mouse brain than cells in the tibia.

Autism is ~4 times more common in males. Jung et al. reveal male-preponderant abnormal behaviors in mice lacking CHD8, a chromatin remodeler, accompanying sexually dimorphic changes in neuronal firing, synaptic transmission, and gene expression.

Controlling urination is a fundamental social need. Keller et al. develop a voluntary urination assay and uncover a subset of brainstem neurons that relax the urethral sphincter, providing insight into urinary control and its dysfunction.

This study shows that excitatory neurons in the ventral hippocampus that express the serotonin receptor 2c gene control attention through direct innervation of the mouse equivalent of the Edinger–Westphal nucleus.

Burt et al. analyze patterns of gene expression across human cortex and show expression primarily varies along a sensory-association hierarchy captured by noninvasive neuroimaging, suggesting an organizing principle for microcircuit specialization.

Poulin et al. developed intersectional genetic approaches to target dopamine neuron subtypes defined by combinatorial gene expression. They demonstrate that dopamine neuron subtypes display distinctive projection patterns to forebrain regions.

The authors introduce a variant GCaMP that is targeted to axons. Due to its strong brightness, signal-to-noise ratio, and photostability, this tool, named axon-GCaMP, enables robust Ca²⁺ imaging in individual axons, including in vivo in awake mice.

Using a deep learning approach to track user-defined body parts during various behaviors across multiple species, the authors show that their toolbox, called DeepLabCut, can achieve human accuracy with only a few hundred frames of training data.