A deep learning based method for large-scale classification, registration, and clustering of in-situ hybridization experiments in the mouse olfactory bulb.

BACKGROUND: The Allen Mouse Brain Atlas allows study of the brain's molecular anatomy at cellular scale, for thousands genes. To fully leverage this resource, one must register histological images of brain tissue - a task made challenging by the brain's structural complexity and heterogeneity, as well as inter-experiment variability. NEW METHOD: We have developed a deep-learning based methodology for classification and registration of thousands of sections of brain tissue, using the mouse olfactory bulb (OB) as a case study. RESULTS: We trained a convolutional neural network (CNN) to derive an image similarity measure for in-situ hybridization experiments, and embedded these in a low-dimensional feature space to guide the design of registration templates. We then compiled a high quality, registered atlas of gene expression for the OB (the first such atlas for the OB, to our knowledge). As proof-of-principle, the atlas was clustered using non-negative matrix factorization to reveal canonical expression motifs, and to identify novel, lamina-specific marker genes. COMPARISON WITH EXISTING METHODS: Our method leverages virtues of CNNs for a set of important problems in molecular neuroanatomy, with performance comparable to existing methods. CONCLUSION: The atlas we have complied allows for intra- and inter-laminar comparisons of gene expression patterns in the OB across thousands of genes, as well identification of canonical expression profiles through clustering. We anticipate that this will be a useful resource for investigators studying the bulb's development and functional topography. Our methods are publicly available for those interested in extending them to other brain areas.

Genome-scale investigation of olfactory system spatial heterogeneity.

The early olfactory system is organized in parallel, with numerous, specialized subsystems established by the modular and topographic projections of sensory inputs. While these anatomical sub-systems are in many cases demarcated by well-known marker genes, we stand to learn considerably more about their possible functional specializations from comprehensive, genome-scale descriptions of their molecular anatomy. Here, we leverage the resources of the Allen Brain Atlas (ABA)-a spatially registered compendium of gene expression for the mouse brain-to investigate the early olfactory system's genomic anatomy. We cluster thousands of genes across thousands of voxels in the ABA to derive several novel parcellations of the olfactory system, and concomitantly discover novel sets of enriched, subregion-specific genes that can serve as a starting point for future inquiry.

Neuro at the Nanoscale: Diffraction-Unlimited Imaging with STED Nanoscopy.

Recent breakthroughs in fluorescence microscopy have pushed spatial resolution well beyond the classical limit imposed by diffraction. As a result, the field of nanoscopy has emerged, and diffraction-unlimited resolution is becoming increasingly common in biomedical imaging applications. In this review, we recap the principles behind STED nanoscopy that allow imaging beyond the diffraction limit, and highlight both historical and recent advances made in the field of neuroscience as a result of this technology.

The number of olfactory stimuli that humans can discriminate is still unknown.

It was recently proposed (Bushdid et al., 2014) that humans can discriminate between at least a trillion olfactory stimuli. Here we show that this claim is the result of a fragile estimation framework capable of producing nearly any result from the reported data, including values tens of orders of magnitude larger or smaller than the one originally reported in (Bushdid et al., 2014). Additionally, the formula used to derive this estimate is well-known to provide an upper bound, not a lower bound as reported. That is to say, the actual claim supported by the calculation is in fact that humans can discriminate at most one trillion olfactory stimuli. We conclude that there is no evidence for the original claim.

Olfaction, valuation, and action: reorienting perception.

In the philosophy of perception, olfaction is the perennial problem child, presenting a range of difficulties to those seeking to define its proper referents, and its phenomenological content. Here, we argue that many of these difficulties can be resolved by recognizing the object-like representation of odors in the brain, and by postulating that the basic objects of olfaction are best defined by their biological value to the organism, rather than physicochemical dimensions of stimuli. Building on this organism-centered account, we speculate that the phenomenological space of olfaction is organized into a number of coarse affective dimensions that apply categorically. This organization may be especially useful for coupling sensation to decision making and instrumental action in a sensory modality where the stimulus space is especially complex and high dimensional.

Categorical dimensions of human odor descriptor space revealed by non-negative matrix factorization.

In contrast to most other sensory modalities, the basic perceptual dimensions of olfaction remain unclear. Here, we use non-negative matrix factorization (NMF)--a dimensionality reduction technique--to uncover structure in a panel of odor profiles, with each odor defined as a point in multi-dimensional descriptor space. The properties of NMF are favorable for the analysis of such lexical and perceptual data, and lead to a high-dimensional account of odor space. We further provide evidence that odor dimensions apply categorically. That is, odor space is not occupied homogenously, but rather in a discrete and intrinsically clustered manner. We discuss the potential implications of these results for the neural coding of odors, as well as for developing classifiers on larger datasets that may be useful for predicting perceptual qualities from chemical structures.

Functional polarity in neurons: what can we learn from studying an exception?

Dendrites and axons typically handle very different aspects of neuronal signaling. However, many of the functional distinctions between these two types of processes are absent in neurons with release-competent dendrites. This raises fundamental questions about the molecular mechanisms that promote and permit functional specialization, and suggests that the 'exceptional' case of presynaptic dendrites may provide important clues on how neuronal polarity is established. To help stimulate thinking on this new front, we summarize some key aspects of the physiology of dendritic neurotransmitter release, together with recent work on the molecular basis of neuronal polarity.

Subthreshold glutamate release from mitral cell dendrites.

The dendrites of a number of neuron types function as presynaptic structures, releasing transmitter after action potentials and dendritic spikes. In this regard, dendrites can function like axons, producing discrete outputs after suprathreshold electrical events. However, as the major site of synaptic inputs, dendrites experience ongoing subthreshold fluctuations in membrane potential, raising the question of whether these subthreshold changes can cause changes in transmitter release. Here, we show that mitral cells of the accessory olfactory bulb release glutamate from their dendrites in response to both subthreshold and suprathreshold stimuli. Whereas subthreshold output was typically low under control conditions, it could be enhanced several fold by pharmacological or endogenous activation of group I metabotropic glutamate receptors. These results indicate that presynaptic dendrites can support two distinct forms of output, and can dynamically regulate how electrical activity is coupled to transmitter release.

Recurrent dendrodendritic inhibition of accessory olfactory bulb mitral cells requires activation of group I metabotropic glutamate receptors.

Metabotropic glutamate receptors (mGluRs) modulate neural excitability and network tone in many brain regions. Expression of mGluRs is particularly high in the accessory olfactory bulb (AOB), a CNS structure critical for detecting chemicals that identify kin and conspecifics. Because of its relative simplicity and its direct projection to the hypothalamus, the AOB provides a model system for studying how mGluRs affect the flow of encoded sensory information to downstream areas. We investigated the role of group I mGluRs in synaptic processing in AOB slices and found that under control conditions, recurrent inhibition of principal neurons (mitral cells) was completely eliminated by the mGluR1 antagonist LY367385 [(S)-(+)-alpha-amino-4-carboxy-2 methylbenzeneacetic acid]. In addition, the group I mGluR agonist DHPG [(S)-3,5-dihydroxyphenylglycine; 20 microM] induced a dramatic increase in the rate of spontaneous IPSCs. This increase was dependent on voltage-gated calcium channels but persisted even after blockade of ionotropic glutamatergic transmission and sodium channels. Together, these results indicate that mGluR1 plays a critical role in controlling information flow through the AOB and suggest that mGluR1 may be an important locus for experience-dependent changes in synaptic function.

Tuft calcium spikes in accessory olfactory bulb mitral cells.

The mammalian accessory olfactory system is critical for the detection and identification of pheromones and the representation of complex stimuli including sex, genetic relatedness, and individual identity. Mitral cells, the principal cells of the accessory olfactory bulb (AOB), receive monosynaptic input from the sensory periphery and already show highly specific response properties, firing selectively for combinations of genetic markers and gender-specific cues. Vomeronasal sensory neuron axons form synapses onto distal tuft-like branches of mitral cell primary dendrites. We have studied dendritic excitability and synaptic integration in AOB mitral cell dendrites, and we show that dendrites of accessory olfactory bulb mitral cells support action potential propagation and can fire regenerative spike-like events that are likely to contribute to the integration of inputs to these cells. These tuft spikes may be important for the specificity of AOB mitral cell responses.