The need for automated and efficient systems for tracking full animal pose has increased with the complexity of behavioral data and analyses. Here we introduce LEAP (LEAP estimates animal pose), a deep-learning-based method for predicting the positions of animal body parts. This framework consists of a graphical interface for labeling of body parts and training the network. LEAP offers fast prediction on new data, and training with as few as 100 frames results in 95% of peak performance. We validated LEAP using videos of freely behaving fruit flies and tracked 32 distinct points to describe the pose of the head, body, wings and legs, with an error rate of <3% of body length. We recapitulated reported findings on insect gait dynamics and demonstrated LEAP’s applicability for unsupervised behavioral classification. Finally, we extended the method to more challenging imaging situations and videos of freely moving mice.

Recent work quantifying postural dynamics has attempted to define the repertoire of behaviors across the entire data set, we use a technique we refer to as cluster sampling. A simple random 117 subset of the movie images are grouped via k-means clustering and then these images are 118 sampled uniformly across groups for labeling. The grouping is based on linear correlations 119 between pixel intensities in the images as a proxy measure for similarity in body pose. The 120 diversity of poses represented using this method can be observed in the centroids of each of the 121 clusters identified (Supplementary Fig. 2). 122 123 Poses in each training image are labeled using a custom GUI with draggable body part markers 124 that form a skeleton (Fig. 1b). For the fruit fly, we track four points on each of the six legs, two 125 points on the wing tips, three points on the thorax and abdomen, and three points on the head 126 for a total of 32 points in every frame. These points were chosen to align with known Drosophila 127 body joints (Supplementary Fig. 3). For every training image, the user drags each skeleton 128 point to the appropriate body part and the program saves the label positions into a self-129 contained file. To enhance the size of the training image set further without the need for hand 130 labeling more frames, we augment the dataset by applying small random rotations and body-131 axis reflections to generate new samples from the labeled data. As the neural network 132 processes the raw images, the rotated and reflected images add new information that the 133 network can use during training. 134 135

No abstract

Chronic stress can have lasting adverse consequences in some individuals, yet others are resilient to the same stressor1,2. While previous work found differences in the intrinsic properties of mesolimbic dopamine (DA) neurons in susceptible and resilient individuals after stress was over;3-10 the causal links between DA activity during stress, dynamic stress-evoked behavior, and individual differences in susceptibility and resilience are not known. Here, we record behavior and neural activity in DA projections to the nucleus accumbens (NAc, signals reward11-14) and to the tail striatum (TS, signals threat15-18) during a multiday chronic social defeat paradigm and discover behavioral and neural signatures of resilience. Using supervised and unsupervised behavioral quantification, we find that resilient and susceptible individuals employ different behavioral strategies during stress. In addition, NAc-DA (but not TS-DA) activity is higher in the proximity of the aggressor in resilient mice, consistent with a greater subjective value of the aggressor. Moreover, NAc-DA tends to be elevated at the onset of fighting back in resilient mice and at the offset of attacks in susceptible mice. To test whether DA activation during defeat can generate resilience, and if its timing with respect to behavior is critical, we performed optogenetic stimulation of NAc-DA in open-loop (randomly timed) during defeat or timed to specific behaviors using real-time pose-tracking and behavioral classification. We find that both open-loop DA activation and fighting-back-timed activation promote resilience, in both cases reorganizing behavior during defeat toward resilience-associated patterns. Attack offset-timed activation promotes avoidance during defeat but does not promote susceptibility afterwards. Together, these data suggest a model whereby, during stress, DA in the NAc can increase resilience primarily by elevating the subjective value of the stressor rather than by reinforcing particular stress-responsive behaviors.