For goal-directed behaviour it is critical that we can both select the appropriate action and learn to modify the underlying movements (for example, the pitch of a note or velocity of a reach) to improve outcomes. The basal ganglia are a critical nexus where circuits necessary for the production of behaviour, such as the neocortex and thalamus, are integrated with reward signalling to reinforce successful, purposive actions. The dorsal striatum, a major input structure of basal ganglia, is composed of two opponent pathways, direct and indirect, thought to select actions that elicit positive outcomes and suppress actions that do not, respectively. Activity-dependent plasticity modulated by reward is thought to be sufficient for selecting actions in the striatum. Although perturbations of basal ganglia function produce profound changes in movement, it remains unknown whether activity-dependent plasticity is sufficient to produce learned changes in movement kinematics, such as velocity. Here we use cell-type-specific stimulation in mice delivered in closed loop during movement to demonstrate that activity in either the direct or indirect pathway is sufficient to produce specific and sustained increases or decreases in velocity, without affecting action selection or motivation. These behavioural changes were a form of learning that accumulated over trials, persisted after the cessation of stimulation, and were abolished in the presence of dopamine antagonists. Our results reveal that the direct and indirect pathways can each bidirectionally control movement velocity, demonstrating unprecedented specificity and flexibility in the control of volition by the basal ganglia.

Deep learning describes a class of machine learning algorithms that are capable of combining raw inputs into layers of intermediate features. These algorithms have recently shown impressive results across a variety of domains. Biology and medicine are data-rich disciplines, but the data are complex and often ill-understood. Hence, deep learning techniques may be particularly well suited to solve problems of these fields. We examine applications of deep learning to a variety of biomedical problems—patient classification, fundamental biological processes and treatment of patients—and discuss whether deep learning will be able to transform these tasks or if the biomedical sphere poses unique challenges. Following from an extensive literature review, we find that deep learning has yet to revolutionize biomedicine or definitively resolve any of the most pressing challenges in the field, but promising advances have been made on the prior state of the art. Even though improvements over previous baselines have been modest in general, the recent progress indicates that deep learning methods will provide valuable means for speeding up or aiding human investigation. Though progress has been made linking a specific neural network's prediction to input features, understanding how users should interpret these models to make testable hypotheses about the system under study remains an open challenge. Furthermore, the limited amount of labelled data for training presents problems in some domains, as do legal and privacy constraints on work with sensitive health records. Nonetheless, we foresee deep learning enabling changes at both bench and bedside with the potential to transform several areas of biology and medicine.

Major resources are now available to develop tools and technologies aimed at dissecting the circuitry and computations underlying behavior, unraveling the underpinnings of brain disorders, and understanding the neural substrates of cognition. Scientists from around the world shared their views around new tools and technologies to drive advances in neuroscience.

Neural stem cells show age-dependent developmental potentials, as evidenced by their production of distinct neuron types at different developmental times. Drosophila neuroblasts produce long, stereotyped lineages of neurons. We searched for factors that could regulate neural temporal fate by RNA-sequencing lineage-specific neuroblasts at various developmental times. We found that two RNA-binding proteins, IGF-II mRNA-binding protein (Imp) and Syncrip (Syp), display opposing high-to-low and low-to-high temporal gradients with lineage-specific temporal dynamics. Imp and Syp promote early and late fates, respectively, in both a slowly progressing and a rapidly changing lineage. Imp and Syp control neuronal fates in the mushroom body lineages by regulating the temporal transcription factor Chinmo translation. Together, the opposing Imp/Syp gradients encode stem cell age, specifying multiple cell fates within a lineage.

Two-photon excitation fluorescence microscopy has revolutionized our understanding of brain structure and function through the high resolution and large penetration depth it offers. Investigating neural structures in vivo requires gaining optical access to the brain, which is typically achieved by replacing a part of the skull with one or several layers of cover glass windows. To compensate for the spherical aberrations caused by the presence of these layers of glass, collar-correction objectives are typically used. However, the efficiency of this correction has been shown to depend significantly on the tilt angle between the glass window surface and the optical axis of the imaging system. Here, we first expand these observations and characterize the effect of the tilt angle on the collected fluorescence signal with thicker windows (double cover slide) and compare these results with an objective devoid of collar-correction. Second, we present a simple optical alignment device designed to rapidly minimize the tilt angle in vivo and align the optical axis of the microscope perpendicularly to the glass window to an angle below 0.25°, thereby significantly improving the imaging quality. Finally, we describe a tilt-correction procedure for users in an in vivo setting, enabling the accurate alignment with a resolution of <0.2° in only few iterations.

Electrophysiology and optical indicators have been used in vertebrate systems to investigate excitable cell firing and calcium transients, but both techniques have been difficult to apply in organisms with powerful reverse genetics. To overcome this limitation, we expressed cameleon proteins, genetically encoded calcium indicators, in the pharyngeal muscle of the nematode worm Caenorhabditis elegans. In intact transgenic animals expressing cameleons, fluorescence ratio changes accompanied muscular contraction, verifying detection of calcium transients. By comparing the magnitude and duration of calcium influx in wild-type and mutant animals, we were able to determine the effects of calcium channel proteins on pharyngeal calcium transients. We also successfully used cameleons to detect electrically evoked calcium transients in individual C. elegans neurons. This technique therefore should have broad applications in analyzing the regulation of excitable cell activity in genetically tractable organisms.

Receptor tyrosine kinases (RTK) are important cell surface receptors that transduce extracellular signals across the plasma membrane. The traditional view of how these receptors function is that ligand binding to the extracellular domains acts as a master-switch that enables receptor monomers to dimerize and subsequently trans-phosphorylate each other on their intracellular domains. However, a growing body of evidence suggests that receptor oligomerization is not merely a consequence of ligand binding, but is instead part of a complex process responsible for regulation of receptor activation. Importantly, the oligomerization dynamics and subsequent activation of these receptors are affected by other cellular components, such as cytoskeletal machineries and cell membrane lipid characteristics. Thus receptor activation is not an isolated molecular event mediated by the ligand-receptor interaction, but instead involves orchestrated interactions between the receptors and other cellular components. Measuring receptor oligomerization dynamics on live cells can yield important insights into the characteristics of these interactions. Therefore, it is imperative to develop techniques that can probe receptor movements on the plasma membrane with optimal temporal and spatial resolutions. Various microscopic techniques have been used for this purpose. Optical techniques including single molecule tracking (SMT) and fluorescence correlation spectroscopy (FCS) measure receptor diffusion on live cells. Receptor-receptor interactions can also be assessed by detecting Förster resonance energy transfer (FRET) between fluorescently-labeled receptors situated in close proximity or by counting the number of receptors within a diffraction limited fluorescence spot (stepwise bleaching). This review will describe recent developments of optical techniques that have been used to study receptor oligomerization on living cells. This article is part of a Special Issue entitled: Interactions between membrane receptors in cellular membranes edited by Kalina Hristova.

With amino acids as model systems, optically active sum frequency generation (OA-SFG) was used to probe the chirality of molecules with a chiral center and an intrinsically achiral chromophore in isotropic solution for the first time. Like that of circular dichroism (CD), the OA-SFG’s near electronic resonance originates from the extrachromophoric chiral perturbation on the carboxyl chromophore. The difference in the relative strengths of OA-SFG and CD among different amino acids can be explained by the difference in the details of perturbations.

Experimental investigations have revealed that synapses possess interesting and, in some cases, unexpected properties. We propose a theoretical framework that accounts for three of these properties: typical central synapses are noisy, the distribution of synaptic weights among central synapses is wide, and synaptic connectivity between neurons is sparse. We also comment on the possibility that synaptic weights may vary in discrete steps. Our approach is based on maximizing information storage capacity of neural tissue under resource constraints. Based on previous experimental and theoretical work, we use volume as a limited resource and utilize the empirical relationship between volume and synaptic weight. Solutions of our constrained optimization problems are not only consistent with existing experimental measurements but also make nontrivial predictions.

The high noise level found in single-particle electron cryo-microscopy (cryo-EM) image data presents a special challenge for three-dimensional (3D) reconstruction of the imaged molecules. The spectral signal-to-noise ratio (SSNR) and related Fourier shell correlation (FSC) functions are commonly used to assess and mitigate the noise-generated error in the reconstruction. Calculation of the SSNR and FSC usually includes the noise in the solvent region surrounding the particle and therefore does not accurately reflect the signal in the particle density itself. Here we show that the SSNR in a reconstructed 3D particle map is linearly proportional to the fractional volume occupied by the particle. Using this relationship, we devise a novel filter (the "single-particle Wiener filter") to minimize the error in a reconstructed particle map, if the particle volume is known. Moreover, we show how to approximate this filter even when the volume of the particle is not known, by optimizing the signal within a representative interior region of the particle. We show that the new filter improves on previously proposed error-reduction schemes, including the conventional Wiener filter as well as figure-of-merit weighting, and quantify the relationship between all of these methods by theoretical analysis as well as numeric evaluation of both simulated and experimentally collected data. The single-particle Wiener filter is applicable across a broad range of existing 3D reconstruction techniques, but is particularly well suited to the Fourier inversion method, leading to an efficient and accurate implementation.