Highly Accurate and Precise Determination of Mouse Mass Using Computer Vision.

Changes in body mass are a key indicator of health and disease in humans and model organisms. Animal body mass is routinely monitored in husbandry and preclinical studies. In rodent studies, the current best method requires manually weighing the animal on a balance which has at least two consequences. First, direct handling of the animal induces stress and can have confounding effects on studies. Second, the acquired mass is static and not amenable to continuous assessment, and rapid mass changes can be missed. A noninvasive and continuous method of monitoring animal mass would have utility in multiple areas of biomedical research. Here, we test the feasibility of determining mouse body mass using video data. We combine computer vision methods with statistical modeling to demonstrate the feasibility of our approach. Our methods determine mouse mass with 4.8% error across highly genetically diverse mouse strains, with varied coat colors and mass. This error is low enough to replace manual weighing with image-based assessment in most mouse studies. We conclude that visual determination of rodent mass using video enables noninvasive and continuous monitoring and can improve animal welfare and preclinical studies.

Stride-level analysis of mouse open field behavior using deep-learning-based pose estimation.

Gait and posture are often perturbed in many neurological, neuromuscular, and neuropsychiatric conditions. Rodents provide a tractable model for elucidating disease mechanisms and interventions. Here, we develop a neural-network-based assay that adopts the commonly used open field apparatus for mouse gait and posture analysis. We quantitate both with high precision across 62 strains of mice. We characterize four mutants with known gait deficits and demonstrate that multiple autism spectrum disorder (ASD) models show gait and posture deficits, implying this is a general feature of ASD. Mouse gait and posture measures are highly heritable and fall into three distinct classes. We conduct a genome-wide association study to define the genetic architecture of stride-level mouse movement in the open field. We provide a method for gait and posture extraction from the open field and one of the largest laboratory mouse gait and posture data resources for the research community.

High-throughput visual assessment of sleep stages in mice using machine learning.

STUDY OBJECTIVES: Sleep is an important biological process that is perturbed in numerous diseases, and assessment of its substages currently requires implantation of electrodes to carry out electroencephalogram/electromyogram (EEG/EMG) analysis. Although accurate, this method comes at a high cost of invasive surgery and experts trained to score EEG/EMG data. Here, we leverage modern computer vision methods to directly classify sleep substages from video data. This bypasses the need for surgery and expert scoring, provides a path to high-throughput studies of sleep in mice. METHODS: We collected synchronized high-resolution video and EEG/EMG data in 16 male C57BL/6J mice. We extracted features from the video that are time and frequency-based and used the human expert-scored EEG/EMG data to train a visual classifier. We investigated several classifiers and data augmentation methods. RESULTS: Our visual sleep classifier proved to be highly accurate in classifying wake, non-rapid eye movement sleep (NREM), and rapid eye movement sleep (REM) states, and achieves an overall accuracy of 0.92 ± 0.05 (mean ± SD). We discover and genetically validate video features that correlate with breathing rates, and show low and high variability in NREM and REM sleep, respectively. Finally, we apply our methods to noninvasively detect that sleep stage disturbances induced by amphetamine administration. CONCLUSIONS: We conclude that machine learning-based visual classification of sleep is a viable alternative to EEG/EMG based scoring. Our results will enable noninvasive high-throughput sleep studies and will greatly reduce the barrier to screening mutant mice for abnormalities in sleep.

Action detection using a neural network elucidates the genetics of mouse grooming behavior.

Automated detection of complex animal behaviors remains a challenging problem in neuroscience, particularly for behaviors that consist of disparate sequential motions. Grooming is a prototypical stereotyped behavior that is often used as an endophenotype in psychiatric genetics. Here, we used mouse grooming behavior as an example and developed a general purpose neural network architecture capable of dynamic action detection at human observer-level performance and operating across dozens of mouse strains with high visual diversity. We provide insights into the amount of human annotated training data that are needed to achieve such performance. We surveyed grooming behavior in the open field in 2457 mice across 62 strains, determined its heritable components, conducted GWAS to outline its genetic architecture, and performed PheWAS to link human psychiatric traits through shared underlying genetics. Our general machine learning solution that automatically classifies complex behaviors in large datasets will facilitate systematic studies of behavioral mechanisms.

Behavior is one of the ultimate and most complex outputs of the body's central nervous system, which controls movement, emotion and mood. It is also influenced by a person's genetics. Scientists studying the link between behavior and genetics often conduct experiments using animals, whose actions can be more easily characterized than humans. However, this involves recording hours of video footage, typically of mice or flies. Researchers must then add labels to this footage, identifying certain behaviors before further analysis. This task of annotating video clips ­ similar to image captioning ­ is very time-consuming for investigators. But it could be automated by applying machine learning algorithms, trained with sufficient data. Some computer programs are already in use to detect patterns of behavior, however, there are some limitations. These programs could detect animal behavior (of flies and mice) in trimmed video clips, but not raw footage, and could not always accommodate different lighting conditions or experimental setups. Here, Geuther et al. set out to improve on these previous efforts to automate video annotation. To do so, they used over 1,250 video clips annotated by experienced researchers to develop a general-purpose neural network for detecting mouse behaviors. After sufficient training, the computer model could detect mouse grooming behaviors in raw, untrimmed video clips just as well as human observers could. It also worked with mice of different coat colors, body shapes and sizes in open field animal tests. Using the new computer model, Geuther et al. also studied the genetics underpinning behavior ­ far more thoroughly than previously possible ­ to explain why mice display different grooming behaviors. The algorithm analyzed 2,250 hours of video featuring over 60 kinds of mice and thousands of other animals. It found that mice bred in the laboratory groom less than mice recently collected from the wild do. Further analyses also identified genes linked to grooming traits in mice and found related genes in humans associated with behavioral disorders. Automating video annotation using machine learning models could alleviate the costs of running lengthy behavioral experiments and enhance the reproducibility of study results. The latter is vital for translating behavioral research findings in mice to humans. This study has also provided insights into the amount of human-annotated training data needed to develop high-performing computer models, along with new understandings of how genetics shapes behavior.

Data-driven statistical modeling of the emergent behavior of biohybrid microrobots.

Multi-agent biohybrid microrobotic systems, owing to their small size and distributed nature, offer powerful solutions to challenges in biomedicine, bioremediation, and biosensing. Synthetic biology enables programmed emergent behaviors in the biotic component of biohybrid machines, expounding vast potential benefits for building biohybrid swarms with sophisticated control schemes. The design of synthetic genetic circuits tailored toward specific performance characteristics is an iterative process that relies on experimental characterization of spatially homogeneous engineered cell suspensions. However, biohybrid systems often distribute heterogeneously in complex environments, which will alter circuit performance. Thus, there is a critically unmet need for simple predictive models that describe emergent behaviors of biohybrid systems to inform synthetic gene circuit design. Here, we report a data-driven statistical model for computationally efficient recapitulation of the motility dynamics of two types of Escherichia coli bacteria-based biohybrid swarms-NanoBEADS and BacteriaBots. The statistical model was coupled with a computational model of cooperative gene expression, known as quorum sensing (QS). We determined differences in timescales for programmed emergent behavior in BacteriaBots and NanoBEADS swarms, using bacteria as a comparative baseline. We show that agent localization and genetic circuit sensitivity strongly influence the timeframe and the robustness of the emergent behavior in both systems. Finally, we use our model to design a QS-based decentralized control scheme wherein agents make independent decisions based on their interaction with other agents and the local environment. We show that synergistic integration of synthetic biology and predictive modeling is requisite for the efficient development of biohybrid systems with robust emergent behaviors.

Robust mouse tracking in complex environments using neural networks.

The ability to track animals accurately is critical for behavioral experiments. For video-based assays, this is often accomplished by manipulating environmental conditions to increase contrast between the animal and the background in order to achieve proper foreground/background detection (segmentation). Modifying environmental conditions for experimental scalability opposes ethological relevance. The biobehavioral research community needs methods to monitor behaviors over long periods of time, under dynamic environmental conditions, and in animals that are genetically and behaviorally heterogeneous. To address this need, we applied a state-of-the-art neural network-based tracker for single mice. We compare three different neural network architectures across visually diverse mice and different environmental conditions. We find that an encoder-decoder segmentation neural network achieves high accuracy and speed with minimal training data. Furthermore, we provide a labeling interface, labeled training data, tuned hyperparameters, and a pretrained network for the behavior and neuroscience communities.

Quantitative Investigation of the Role of Intra-/Intercellular Dynamics in Bacterial Quorum Sensing.

Bacteria utilize diffusible signals to regulate population density-dependent coordinated gene expression in a process called quorum sensing (QS). While the intracellular regulatory mechanisms of QS are well-understood, the effect of spatiotemporal changes in the population configuration on the sensitivity and robustness of the QS response remains largely unexplored. Using a microfluidic device, we quantitatively characterized the emergent behavior of a population of swimming E. coli bacteria engineered with the lux QS system and a GFP reporter. We show that the QS activation time follows a power law with respect to bacterial population density, but this trend is disrupted significantly by microscale variations in population configuration and genetic circuit noise. We then developed a computational model that integrates population dynamics with genetic circuit dynamics to enable accurate (less than 7% error) quantitation of the bacterial QS activation time. Through modeling and experimental analyses, we show that changes in spatial configuration of swimming bacteria can drastically alter the QS activation time, by up to 22%. The integrative model developed herein also enables examination of the performance robustness of synthetic circuits with respect to growth rate, circuit sensitivity, and the population's initial size and spatial structure. Our framework facilitates quantitative tuning of microbial systems performance through rational engineering of synthetic ribosomal binding sites. We have demonstrated this through modulation of QS activation time over an order of magnitude. Altogether, we conclude that predictive engineering of QS-based bacterial systems requires not only the precise temporal modulation of gene expression (intracellular dynamics) but also accounting for the spatiotemporal changes in population configuration (intercellular dynamics).