Motor neurons generate pose-targeted movements via proprioceptive sculpting.

Motor neurons are the final common pathway1 through which the brain controls movement of the body, forming the basic elements from which all movement is composed. Yet how a single motor neuron contributes to control during natural movement remains unclear. Here we anatomically and functionally characterize the individual roles of the motor neurons that control head movement in the fly, Drosophila melanogaster. Counterintuitively, we find that activity in a single motor neuron rotates the head in different directions, depending on the starting posture of the head, such that the head converges towards a pose determined by the identity of the stimulated motor neuron. A feedback model predicts that this convergent behaviour results from motor neuron drive interacting with proprioceptive feedback. We identify and genetically2 suppress a single class of proprioceptive neuron3 that changes the motor neuron-induced convergence as predicted by the feedback model. These data suggest a framework for how the brain controls movements: instead of directly generating movement in a given direction by activating a fixed set of motor neurons, the brain controls movements by adding bias to a continuing proprioceptive-motor loop.

Optimization of avian perching manoeuvres.

Perching at speed is among the most demanding flight behaviours that birds perform1,2 and is beyond the capability of most autonomous vehicles. Smaller birds may touch down by hovering3-8, but larger birds typically swoop up to perch1,2-presumably because the adverse scaling of their power margin prohibits hovering9 and because swooping upwards transfers kinetic to potential energy before collision1,2,10. Perching demands precise control of velocity and pose11-14, particularly in larger birds for which scale effects make collisions especially hazardous6,15. However, whereas cruising behaviours such as migration and commuting typically minimize the cost of transport or time of flight16, the optimization of such unsteady flight manoeuvres remains largely unexplored7,17. Here we show that the swooping trajectories of perching Harris' hawks (Parabuteo unicinctus) minimize neither time nor energy alone, but rather minimize the distance flown after stalling. By combining motion capture data from 1,576 flights with flight dynamics modelling, we find that the birds' choice of where to transition from powered dive to unpowered climb minimizes the distance over which high lift coefficients are required. Time and energy are therefore invested to provide the control authority needed to glide safely to the perch, rather than being minimized directly as in technical implementations of autonomous perching under nonlinear feedback control12 and deep reinforcement learning18,19. Naive birds learn this behaviour on the fly, so our findings suggest a heuristic principle that could guide reinforcement learning of autonomous perching.

A brainstem map for visceral sensations.

The nervous system uses various coding strategies to process sensory inputs. For example, the olfactory system uses large receptor repertoires and is wired to recognize diverse odours, whereas the visual system provides high acuity of object position, form and movement1-5. Compared to external sensory systems, principles that underlie sensory processing by the interoceptive nervous system remain poorly defined. Here we developed a two-photon calcium imaging preparation to understand internal organ representations in the nucleus of the solitary tract (NTS), a sensory gateway in the brainstem that receives vagal and other inputs from the body. Focusing on gut and upper airway stimuli, we observed that individual NTS neurons are tuned to detect signals from particular organs and are topographically organized on the basis of body position. Moreover, some mechanosensory and chemosensory inputs from the same organ converge centrally. Sensory inputs engage specific NTS domains with defined locations, each containing heterogeneous cell types. Spatial representations of different organs are further sharpened in the NTS beyond what is achieved by vagal axon sorting alone, as blockade of brainstem inhibition broadens neural tuning and disorganizes visceral representations. These findings reveal basic organizational features used by the brain to process interoceptive inputs.

Lightning Pose: improved animal pose estimation via semi-supervised learning, Bayesian ensembling and cloud-native open-source tools.

Contemporary pose estimation methods enable precise measurements of behavior via supervised deep learning with hand-labeled video frames. Although effective in many cases, the supervised approach requires extensive labeling and often produces outputs that are unreliable for downstream analyses. Here, we introduce 'Lightning Pose', an efficient pose estimation package with three algorithmic contributions. First, in addition to training on a few labeled video frames, we use many unlabeled videos and penalize the network whenever its predictions violate motion continuity, multiple-view geometry and posture plausibility (semi-supervised learning). Second, we introduce a network architecture that resolves occlusions by predicting pose on any given frame using surrounding unlabeled frames. Third, we refine the pose predictions post hoc by combining ensembling and Kalman smoothing. Together, these components render pose trajectories more accurate and scientifically usable. We released a cloud application that allows users to label data, train networks and process new videos directly from the browser.

Different mechanisms of contextual inference govern associatively learned and sensory-evoked postural responses.

Human standing balance relies on the continuous monitoring and integration of sensory signals to infer our body's motion and orientation within the environment. However, when sensory information is no longer contextually relevant to balancing the body (e.g., when sensory and motor signals are incongruent), sensory-evoked balance responses are rapidly suppressed, much earlier than any conscious perception of changes in balance control. Here, we used a robotic balance simulator to assess whether associatively learned postural responses are similarly modulated by sensorimotor incongruence and contextual relevance to postural control. Twenty-nine participants in three groups were classically conditioned to generate postural responses to whole-body perturbations when presented with an initially neutral sound cue. During catch and extinction trials, participants received only the auditory stimulus but in different sensorimotor states corresponding to their group: 1) during normal active balance, 2) while immobilized, and 3) throughout periods where the computer subtly removed active control over balance. In the balancing and immobilized states, conditioned responses were either evoked or suppressed, respectively, according to the (in)ability to control movement. Following the immobilized state, conditioned responses were renewed when balance was restored, indicating that conditioning was retained but only expressed when contextually relevant. In contrast, conditioned responses persisted in the computer-controlled state even though there was no causal relationship between motor and sensory signals. These findings suggest that mechanisms responsible for sensory-evoked and conditioned postural responses do not share a single, central contextual inference and assessment of their relevance to postural control, and may instead operate in parallel.

Natural scenes reveal diverse representations of 2D and 3D body pose in the human brain.

Human pose, defined as the spatial relationships between body parts, carries instrumental information supporting the understanding of motion and action of a person. A substantial body of previous work has identified cortical areas responsive to images of bodies and different body parts. However, the neural basis underlying the visual perception of body part relationships has received less attention. To broaden our understanding of body perception, we analyzed high-resolution fMRI responses to a wide range of poses from over 4,000 complex natural scenes. Using ground-truth annotations and an application of three-dimensional (3D) pose reconstruction algorithms, we compared similarity patterns of cortical activity with similarity patterns built from human pose models with different levels of depth availability and viewpoint dependency. Targeting the challenge of explaining variance in complex natural image responses with interpretable models, we achieved statistically significant correlations between pose models and cortical activity patterns (though performance levels are substantially lower than the noise ceiling). We found that the 3D view-independent pose model, compared with two-dimensional models, better captures the activation from distinct cortical areas, including the right posterior superior temporal sulcus (pSTS). These areas, together with other pose-selective regions in the LOTC, form a broader, distributed cortical network with greater view-tolerance in more anterior patches. We interpret these findings in light of the computational complexity of natural body images, the wide range of visual tasks supported by pose structures, and possible shared principles for view-invariant processing between articulated objects and ordinary, rigid objects.

Spaceflight-Induced Intracranial Hypertension and Visual Impairment: Pathophysiology and Countermeasures.

Visual impairment intracranial pressure (VIIP) syndrome is considered an unexplained major risk for future long-duration spaceflight. NASA recently redefined this syndrome as Spaceflight-Associated Neuro-ocular Syndrome (SANS). Evidence thus reviewed supports that chronic, mildly elevated intracranial pressure (ICP) in space (as opposed to more variable ICP with posture and activity on Earth) is largely accounted for by loss of hydrostatic pressures and altered hemodynamics in the intracranial circulation and the cerebrospinal fluid system. In space, an elevated pressure gradient across the lamina cribrosa, caused by a chronic but mildly elevated ICP, likely elicits adaptations of multiple structures and fluid systems in the eye which manifest themselves as the VIIP syndrome. A chronic mismatch between ICP and intraocular pressure (IOP) in space may acclimate the optic nerve head, lamina cribrosa, and optic nerve subarachnoid space to a condition that is maladaptive to Earth, all contributing to the pathogenesis of space VIIP syndrome. Relevant findings help to evaluate whether artificial gravity is an appropriate countermeasure to prevent this seemingly adverse effect of long-duration spaceflight.

Implanted System for Orthostatic Hypotension in Multiple-System Atrophy.

Orthostatic hypotension is a cardinal feature of multiple-system atrophy. The upright posture provokes syncopal episodes that prevent patients from standing and walking for more than brief periods. We implanted a system to restore regulation of blood pressure and enable a patient with multiple-system atrophy to stand and walk after having lost these abilities because of orthostatic hypotension. This system involved epidural electrical stimulation delivered over the thoracic spinal cord with accelerometers that detected changes in body position. (Funded by the Defitech Foundation.).

Corticocortical and corticomuscular connectivity dynamics in standing posture: electroencephalography study.

Cortical mechanism is necessary for human standing control. Previous research has demonstrated that cortical oscillations and corticospinal excitability respond flexibly to postural demands. However, it is unclear how corticocortical and corticomuscular connectivity changes dynamically during standing with spontaneous postural sway and over time. This study investigated the dynamics of sway- and time-varying connectivity using electroencephalography and electromyography. Electroencephalography and electromyography were recorded in sitting position and 3 standing postures with varying base-of-support: normal standing, one-leg standing, and standing on a piece of wood. For sway-varying connectivity, corticomuscular connectivity was calculated based on the timing of peak velocity in anteroposterior sway. For time-varying connectivity, corticocortical connectivity was measured using the sliding-window approach. This study found that corticomuscular connectivity was strengthened at the peak velocity of postural sway in the γ- and ß-frequency bands. For time-varying corticocortical connectivity, the θ-connectivity in all time-epoch was classified into 7 clusters including posture-relevant component. In one of the 7 clusters, strong connectivity pairs were concentrated in the mid-central region, and the proportion of epochs under narrow-base standing conditions was significantly higher, indicating a functional role for posture balance. These findings shed light on the connectivity dynamics and cortical oscillation that govern standing balance.

Effect of a False Inertial Cue in the Velocity-Storage Circuit on Head Posture and Inertia Perception.

The velocity-storage circuit participates in the vestibulopostural reflex, but its role in the postural reflex requires further elucidation. The velocity-storage circuit differentiates gravitoinertial information into gravitational and inertial cues using rotational cues. This implies that a false rotational cue can cause an erroneous estimation of gravity and inertial cues. We hypothesized the velocity-storage circuit is a common gateway for all vestibular reflex pathways and tested that hypothesis by measuring the postural and perceptual responses from a false inertial cue estimated in the velocity-storage circuit. Twenty healthy human participants (40.5 ± 8.2 years old, 6 men) underwent two different sessions of earth-vertical axis rotations at 120°/s for 60 s. During each session, the participants were rotated clockwise and then counterclockwise with two different starting head positions (head-down and head-up). During the first (control) session, the participants kept a steady head position at the end of rotation. During the second (test) session, the participants changed their head position at the end of rotation, from head-down to head-up or vice versa. The head position and inertial motion perception at the end of rotation were aligned with the inertia direction anticipated by the velocity-storage model. The participants showed a significant correlation between postural and perceptual responses. The velocity-storage circuit appears to be a shared neural integrator for the vestibulopostural reflex and vestibular perception. Because the postural responses depended on the inertial direction, the postural instability in vestibular disorders may be the consequence of the vestibulopostural reflex responding to centrally estimated false vestibular cues.SIGNIFICANCE STATEMENT The velocity-storage circuit appears to participate in the vestibulopostural reflex, which stabilizes the head and body position in space. However, it is still unclear whether the velocity-storage circuit for the postural reflex is in common with that involved in eye movement and perception. We evaluated the postural and perceptual responses to a false inertial cue estimated by the velocity-storage circuit. The postural and perceptual responses were consistent with the inertia direction predicted in the velocity-storage model and were correlated closely with each other. These results show that the velocity-storage circuit is a shared neural integrator for vestibular-driven responses and suggest that the vestibulopostural response to a false vestibular cue is the pathomechanism of postural instability clinically observed in vestibular disorders.

Tilt in Place Microscopy: a Simple, Low-Cost Solution to Image Neural Responses to Body Rotations.

Animals use information about gravity and other destabilizing forces to balance and navigate through their environment. Measuring how brains respond to these forces requires considerable technical knowledge and/or financial resources. We present a simple alternative-Tilt In Place Microscopy (TIPM), a low-cost and noninvasive way to measure neural activity following rapid changes in body orientation. Here, we used TIPM to study vestibulospinal neurons in larval zebrafish during and immediately after roll tilts. Vestibulospinal neurons responded with reliable increases in activity that varied as a function of ipsilateral tilt amplitude. TIPM differentiated tonic (i.e., sustained tilt) from phasic responses, revealing coarse topography of stimulus sensitivity in the lateral vestibular nucleus. Neuronal variability across repeated sessions was minor relative to trial-to-trial variability, allowing us to use TIPM for longitudinal studies of the same neurons across two developmental time points. There, we observed global increases in response strength and systematic changes in the neural representation of stimulus direction. Our data extend classical characterization of the body tilt representation by vestibulospinal neurons and establish the utility of TIPM to study the neural basis of balance, especially in developing animals.SIGNIFICANCE STATEMENT Vestibular sensation influences everything from navigation to interoception. Here, we detail a straightforward, validated, and nearly universal approach to image how the nervous system senses and responds to body tilts. We use our new method to replicate and expand on past findings of tilt sensing by a conserved population of spinal-projecting vestibular neurons. The simplicity and broad compatibility of our approach will democratize the study of the response of the brain to destabilization, particularly across development.

Relationship of device measured physical activity type and posture with cardiometabolic health markers: pooled dose-response associations from the Prospective Physical Activity, Sitting and Sleep Consortium.

AIMS/HYPOTHESIS: The aim of this study was to examine the dose-response associations of device-measured physical activity types and postures (sitting and standing time) with cardiometabolic health. METHODS: We conducted an individual participant harmonised meta-analysis of 12,095 adults (mean ± SD age 54.5±9.6 years; female participants 54.8%) from six cohorts with thigh-worn accelerometry data from the Prospective Physical Activity, Sitting and Sleep (ProPASS) Consortium. Associations of daily walking, stair climbing, running, standing and sitting time with a composite cardiometabolic health score (based on standardised z scores) and individual cardiometabolic markers (BMI, waist circumference, triglycerides, HDL-cholesterol, HbA1c and total cholesterol) were examined cross-sectionally using generalised linear modelling and cubic splines. RESULTS: We observed more favourable composite cardiometabolic health (i.e. z score <0) with approximately 64 min/day walking (z score [95% CI] -0.14 [-0.25, -0.02]) and 5 min/day stair climbing (-0.14 [-0.24, -0.03]). We observed an equivalent magnitude of association at 2.6 h/day standing. Any amount of running was associated with better composite cardiometabolic health. We did not observe an upper limit to the magnitude of the dose-response associations for any activity type or standing. There was an inverse dose-response association between sitting time and composite cardiometabolic health that became markedly less favourable when daily durations exceeded 12.1 h/day. Associations for sitting time were no longer significant after excluding participants with prevalent CVD or medication use. The dose-response pattern was generally consistent between activity and posture types and individual cardiometabolic health markers. CONCLUSIONS/INTERPRETATION: In this first activity type-specific analysis of device-based physical activity, ~64 min/day of walking and ~5.0 min/day of stair climbing were associated with a favourable cardiometabolic risk profile. The deleterious associations of sitting time were fully attenuated after exclusion of participants with prevalent CVD and medication use. Our findings on cardiometabolic health and durations of different activities of daily living and posture may guide future interventions involving lifestyle modification.

Development of autonomic heart rate modulations during childhood and adolescence.

Autonomic control of heart rate is well known in adult subjects, but limited data are available on the development of the heart rate control during childhood and adolescence. Continuous 12-lead electrocardiograms were recorded in 1045 healthy children and adolescents (550 females) aged 4 to 19 years during postural manoeuvres involving repeated 10-min supine, unsupported sitting, and unsupported standing positions. In each position, heart rate was measured, and heart rate variability indices were evaluated (SDNN, RMSSD, and high (HF) and low (LF) frequency components were obtained). Quasi-normalized HF frequency components were defined as qnHF = HF/(HF + LF). These measurements were, among others, related to age using linear regressions. In supine position, heart rate decreases per year of age were significant in both sexes but lower in females than in males. In standing position, these decreases per year of age were substantially lowered. RMSSD and qnHF indices were independent of age in supine position but significantly decreased with age in sitting and standing positions. Correspondingly, LF/HF proportions showed steep increases with age in sitting and standing positions but not in the supine position. The study suggests that baseline supine parasympathetic influence shows little developmental changes during childhood and adolescence but that in young children, sympathetic branch is less responsive to vagal influence. While vagal influences modulate cardiac periods in young and older children equally, they are less able to suppress the sympathetic influence in younger children.

Cortical beta oscillations help synchronise muscles during static posture holding in healthy motor control.

How cortical oscillations are involved in the coordination of functionally coupled muscles and how this is modulated by different movement contexts (static vs dynamic) remains unclear. Here, this is investigated by recording high-density electroencephalography (EEG) and electromyography (EMG) from different forearm muscles while healthy participants (n = 20) performed movement tasks (static and dynamic posture holding, and reaching) with their dominant hand. When dynamic perturbation was applied, beta band (15-35 Hz) activities in the motor cortex contralateral to the performing hand reduced during the holding phase, comparative to when there was no perturbation. During static posture holding, transient periods of increased cortical beta oscillations (beta bursts) were associated with greater corticomuscular coherence and increased phase synchrony between muscles (intermuscular coherence) in the beta frequency band compared to the no-burst period. This effect was not present when resisting dynamic perturbation. The results suggest that cortical beta bursts assist synchronisation of different muscles during static posture holding in healthy motor control, contributing to the maintenance and stabilisation of functional muscle groups. Theoretically, increased cortical beta oscillations could lead to exaggerated synchronisation in different muscles making the initialisation of movements more difficult, as observed in Parkinson's disease.

Responses to brief perturbations of stance: EMG, midline cortical, and subcortical changes.

We studied simultaneous EMG and midline EEG responses, including over the cerebellum, in 10 standing subjects (35 ± 15 yr; 5 females, 5 males). Recordings were made following repeated taps to the sternum, stimuli known to evoke short-latency EMG responses in leg muscles, consistent with postural reflexes. EEG power had relatively more high-frequency components (>30 Hz) when recorded from electrodes over the cerebellum (Iz and SIz) compared with other midline electrodes. We confirmed a previous report using a similar stimulus that evoked short-latency potentials over the cerebellum. We showed clear midline-evoked EEG potentials occurring at short latency over the cerebellum (P23, N31, N42, and P54) and frontally (N28 and N57) before the previously described perturbation-evoked potential (P1/N1/P2). The P23 response correlated with the subsequent EMG response in the tibialis anterior muscles (r = 0.72, P = 0.018), confirming and extending previous observations. We did not find a correlation with the N1 amplitude. We conclude that early activity occurs from electrodes over the inion in response to a brief tap to the sternum. This is likely to represent cerebellar activity and it appears to modulate short-latency postural EMG responses.NEW & NOTEWORTHY We studied the effects of a brief tap to the sternum in human subjects, known to evoke short-latency postural responses. Using an extended EEG recording system, we showed early evoked responses over the midline cerebellum, including the P23 potential, which correlated with the EMG responses in tibialis anterior, consistent with a cerebellar role in postural reflexes. The stimulus also evoked later EEG responses, including the perturbation potential.

Predictive posture stabilization before contact with moving objects: equivalence of smooth pursuit tracking and peripheral vision.

Postural stabilization is essential to effectively interact with our environment. Humans preemptively adjust their posture to counteract impending disturbances, such as those encountered during interactions with moving objects, a phenomenon known as anticipatory postural adjustments (APAs). APAs are thought to be influenced by predictive models that incorporate object motion via retinal motion and extraretinal signals. Building on our previous work that examined APAs in relation to the perceived momentum of moving objects, here we explored the impact of object motion within different visual field sectors on the human capacity to anticipate motion and prepare APAs for contact between virtual moving objects and the limb. Participants interacted with objects moving toward them under different gaze conditions. In one condition, participants fixated on either a central point (central fixation) or left-right of the moving object (peripheral fixation), whereas in another, they followed the moving object with smooth pursuit eye movements (SPEMs). We found that APAs had the smallest magnitude in the central fixation condition and that no notable differences in APAs were apparent between the SPEM and peripheral fixation conditions. This suggests that the visual system can accurately perceive motion of objects in peripheral vision for posture stabilization. Using Bayesian model averaging, we also evaluated the contribution of different gaze variables, such as eye velocity and gain (ratio of eye and object velocity) and showed that both eye velocity and gain signals were significant predictors of APAs. Taken together, our study underscores the roles of oculomotor signals in the modulation of APAs.NEW & NOTEWORTHY We show that the human visuomotor system can detect motion in peripheral vision and make anticipatory adjustments to posture before contact with moving objects, just as effectively as when the eye movement system tracks those objects with smooth pursuit eye movements. These findings pave the way for research into how age-induced changes in spatial vision, eye movements, and motion perception could affect the control of limb movements and postural stability during motion-mediated interactions with objects.

Sitting leg vasculopathy: potential adaptations beyond the endothelium.

Increased sitting time, the most common form of sedentary behavior, is an independent risk factor for all-cause and cardiovascular disease mortality; however, the mechanisms linking sitting to cardiovascular risk remain largely elusive. Studies over the last decade have led to the concept that excessive time spent in the sitting position and the ensuing reduction in leg blood flow-induced shear stress cause endothelial dysfunction. This conclusion has been mainly supported by studies using flow-mediated dilation in the lower extremities as the measured outcome. In this review, we summarize evidence from classic studies and more recent ones that collectively support the notion that prolonged sitting-induced leg vascular dysfunction is likely also attributable to changes occurring in vascular smooth muscle cells (VSMCs). Indeed, we provide evidence that prolonged constriction of resistance arteries can lead to modifications in the structural characteristics of the vascular wall, including polymerization of actin filaments in VSMCs and inward remodeling, and that these changes manifest in a time frame that is consistent with the vascular changes observed with prolonged sitting. We expect this review will stimulate future studies with a focus on VSMC cytoskeletal remodeling as a potential target to prevent the detrimental vascular ramifications of too much sitting.

Differences in heart rate responses to upright posture are associated with variations in the high-frequency power of heart rate variability.

High resting heart rate is a cardiovascular risk factor, but limited data exist on the underlying hemodynamics and reproducibility of supine-to-upright increase in heart rate. We recorded noninvasive hemodynamics in 574 volunteers [age, 44.9 yr; body mass index (BMI), 26.4 kg/m2; 49% male] during passive head-up tilt (HUT) using whole body impedance cardiography and radial artery tonometry. Heart rate regulation was evaluated using heart rate variability (HRV) analyses. Comparisons were made between quartiles of supine-to-upright heart rate changes, in which heart rate at rest ranged 62.6-64.8 beats/min (P = 0.285). The average upright increases in heart rate in the quartiles 1-4 were 4.7, 9.9, 13.5, and 21.0 beats/min, respectively (P < 0.0001). No differences were observed in the low-frequency power of HRV, whether in the supine or upright position, or in the high-frequency power of HRV in the supine position. Upright high-frequency power of HRV was highest in quartile 1 with lowest upright heart rate and lowest in quartile 4 with highest upright heart rate. Mean systolic blood pressure before and during HUT (126 vs. 108 mmHg) and the increase in systemic vascular resistance during HUT (650 vs. 173 dyn·s/cm5/m2) were highest in quartile 1 and lowest in quartile 4. The increases in heart rate during HUT on three separate occasions several weeks apart were highly reproducible (r = 0.682) among 215 participants. To conclude, supine-to-upright increase in heart rate is a reproducible phenotype with underlying differences in the modulation of cardiac parasympathetic tone and systemic vascular resistance. As heart rate at rest influences prognosis, future research should elucidate the prognostic significance of these phenotypic differences.NEW & NOTEWORTHY Subjects with similar supine heart rates are characterized by variable increases in heart rate during upright posture. Individual heart rate increases in response to upright posture are highly reproducible as hemodynamic phenotypes and present underlying differences in the modulation of cardiac parasympathetic tone and systemic vascular resistance. These results indicate that resting heart rate obtained in the supine position alone is not an optimal means of classifying people into groups with differences in cardiovascular function.

A machine learning based method for tracking of simultaneously imaged neural activity and body posture of freely moving maggot.

To understand neural basis of animal behavior, it is necessary to monitor neural activity and behavior in freely moving animal before building relationship between them. Here we use light sheet fluorescence microscope (LSFM) combined with microfluidic chip to simultaneously capture neural activity and body movement in small freely behaving Drosophila larva. We develop a transfer learning based method to simultaneously track the continuously changing body posture and activity of neurons that move together using a sub-region tracking network with a precise landmark estimation network for the inference of target landmark trajectory. Based on the tracking of each labelled neuron, the activity of the neuron indicated by fluorescent intensity is calculated. For each video, annotation of only 20 frames in a video is sufficient to yield human-level accuracy for all other frames. The validity of this method is further confirmed by reproducing the activity pattern of PMSIs (period-positive median segmental interneurons) and larval movement as previously reported. Using this method, we disclosed the correlation between larval movement and left-right asymmetry in activity of a group of unidentified neurons labelled by R52H01-Gal4 and further confirmed the roles of these neurons in bilateral balance of body contraction during larval crawling by genetic inhibition of these neurons. Our method provides a new tool for accurate extraction of neural activities and movement of freely behaving small-size transparent animals.

LiftPose3D, a deep learning-based approach for transforming two-dimensional to three-dimensional poses in laboratory animals.

Markerless three-dimensional (3D) pose estimation has become an indispensable tool for kinematic studies of laboratory animals. Most current methods recover 3D poses by multi-view triangulation of deep network-based two-dimensional (2D) pose estimates. However, triangulation requires multiple synchronized cameras and elaborate calibration protocols that hinder its widespread adoption in laboratory studies. Here we describe LiftPose3D, a deep network-based method that overcomes these barriers by reconstructing 3D poses from a single 2D camera view. We illustrate LiftPose3D's versatility by applying it to multiple experimental systems using flies, mice, rats and macaques, and in circumstances where 3D triangulation is impractical or impossible. Our framework achieves accurate lifting for stereotypical and nonstereotypical behaviors from different camera angles. Thus, LiftPose3D permits high-quality 3D pose estimation in the absence of complex camera arrays and tedious calibration procedures and despite occluded body parts in freely behaving animals.