Sensory systems support generalization by representing features that persist under input variation; however, identifying the neuronal basis of these invariances remains difficult due to high-dimensional and nonlinear neural computations. Here we leverage the inception loop paradigm, iterating between large-scale recordings, predictive models and in silico experiments with in vivo verification, to characterize neuronal invariances in mouse primary visual cortex (V1). We synthesize varied exciting inputs (VEIs), dissimilar images that drive target neurons. These VEIs revealed a new bipartite invariance: one subfield encodes a shift-tolerant high-frequency texture and the other encodes a fixed low-frequency pattern. This division aligns with object boundaries defined by spatial frequency differences in highly activating images, suggesting a contribution to segmentation. Analysis of the MICrONS dataset revealed a hierarchy of excitatory neurons in mouse V1 layers 2/3: postsynaptic neurons exhibited greater invariance than their presynaptic inputs, while neurons with lower invariance formed more connections. Together, these results provide insights and scalable methodology for mapping neuronal invariances.

Unsupervised machine learning is widely used to mine large, unlabelled datasets to make data-driven discoveries in critical domains such as climate science, biomedicine, astronomy, chemistry and more. However, despite its widespread utilization, there is a lack of standardization in unsupervised learning workflows for making reliable and reproducible scientific discoveries. In this paper, we present a structured workflow for using unsupervised learning techniques in science. We highlight and discuss best practices starting with formulating validatable scientific questions, conducting robust data preparation and exploration, using a range of modelling techniques, performing rigorous validation by evaluating the stability and generalizability of unsupervised learning conclusions, and promoting effective communication and documentation of results to ensure reproducible scientific discoveries. To illustrate our proposed workflow, we present a case study from astronomy, seeking to refine globular clusters of Milky Way stars based upon their chemical composition. Our case study highlights the importance of validation and illustrates how the benefits of a carefully designed workflow for unsupervised learning can advance scientific discovery. This article is part of the theme issue 'Statistical workflow'.

Patchwork learning arises as a new and challenging data collection paradigm where both samples and features are observed in fragmented subsets. Due to technological limitations and measurement expenses, such patchwork data structures are frequently seen in applications like neuroscience, healthcare, and genomics, among others. Instead of analyzing each data patch separately, it is highly desirable to extract comprehensive knowledge from the whole dataset. In this work, we focus on the clustering problem in patchwork learning, aiming at discovering clusters among all samples even when some are never jointly observed for any feature. We propose a novel spectral clustering method called Cluster Quilting, consisting of (i) patch ordering that exploits the overlapping structure amongst all patches, (ii) patch-wise SVD, (iii) sequential linear mapping of top singular vectors for patch overlaps, followed by (iv) k-means on the combined and weighted singular vectors. We establish theoretical guarantees via a non-asymptotic misclustering rate bound that reflects both properties of the patch-wise observation regime as well as the clustering signal and noise dependencies. We validate our Cluster Quilting algorithm through empirical studies on both simulated and real datasets, where we show Cluster Quilting yields more accurate and scientifically more plausible clusters than other approaches. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

The complexity of neural circuits makes it challenging to decipher the brain's algorithms of intelligence. Recent breakthroughs in deep learning have produced models that accurately simulate brain activity, enhancing our understanding of the brain's computational objectives and neural coding. However, it is difficult for such models to generalize beyond their training distribution, limiting their utility. The emergence of foundation models1 trained on vast datasets has introduced a new artificial intelligence paradigm with remarkable generalization capabilities. Here we collected large amounts of neural activity from visual cortices of multiple mice and trained a foundation model to accurately predict neuronal responses to arbitrary natural videos. This model generalized to new mice with minimal training and successfully predicted responses across various new stimulus domains, such as coherent motion and noise patterns. Beyond neural response prediction, the model also accurately predicted anatomical cell types, dendritic features and neuronal connectivity within the MICrONS functional connectomics dataset2. Our work is a crucial step towards building foundation models of the brain. As neuroscience accumulates larger, multimodal datasets, foundation models will reveal statistical regularities, enable rapid adaptation to new tasks and accelerate research.

The complexity of neural circuits makes it challenging to decipher the brain's algorithms of intelligence. Recent breakthroughs in deep learning have produced models that accurately simulate brain activity, enhancing our understanding of the brain's computational objectives and neural coding. However, it is difficult for such models to generalize beyond their training distribution, limiting their utility. The emergence of foundation models1 trained on vast datasets has introduced a new artificial intelligence paradigm with remarkable generalization capabilities. Here we collected large amounts of neural activity from visual cortices of multiple mice and trained a foundation model to accurately predict neuronal responses to arbitrary natural videos. This model generalized to new mice with minimal training and successfully predicted responses across various new stimulus domains, such as coherent motion and noise patterns. Beyond neural response prediction, the model also accurately predicted anatomical cell types, dendritic features and neuronal connectivity within the MICrONS functional connectomics dataset2. Our work is a crucial step towards building foundation models of the brain. As neuroscience accumulates larger, multimodal datasets, foundation models will reveal statistical regularities, enable rapid adaptation to new tasks and accelerate research.

Facemask pneumotachography is a common, precise experimental approach for measuring respiratory outcomes, including in neonatal rodents aged postnatal day 7-8. Pneumotachography has been applied in autoresuscitation experiments that inform upon Sudden Infant Death Syndrome (SIDS) where failure of autoresuscitation is hypothesized to be a common endpoint. However, current approaches rely on an experimenter’s direct real-time waveform interpretation and highly variable reaction times to change gas exposures. We have developed Looper, which represents a novel and significant improvement on the typical observer-based interpretation of real-time data to initiate challenge and recovery gas applications. Among the goals of this system are uncovering features, physiological signal derived biomarkers, and risk factors related to neonate cardiorespiratory physiology and autoresuscitation failure. Our platform has also benefited from updates to our previously published Breathe Easy software, where we have expanded the analytical capabilities of the software to measure critical features within a typical autoresuscitation cycle in neonatal mice. Automation of the neonate autoresuscitation reflex assay has created several opportunities to interrogate basic biology, genetic and exposure risks, and potential therapeutics related to autoresuscitation failure and SIDS. Iterative expansion of our automated systems further expand the platform’s potential but requires approaches to manage the pipeline. Currently, initiation of the experiment requires user attention and manual data input. However, workflow improvements including QR code scanning and high-throughput animal handling devices reduce error rates and improve productivity. Version control of software and settings used for data collection have also been implemented to manage tracking of changes to algorithm and default experiment settings. Our previously published Breathe Easy software has also been improved to accept neonate autoresuscitation data and provide specialized analyses specific to this physiological assay. We continue to develop strategies to manage this data collection and analysis pipeline including the use of code repositories and remote system management tools, like ansible, and scheduling and automation of file management and data analysis. Here, we provide an update on the current status of our platform and the impact of refinements that have been implemented. UM1HG006348, R01HL130249, R01HL161142, BCM-CVRI Pilot Grant Funding, Support from BCM's Advanced Technology Cores This abstract was presented at the American Physiology Summit 2025 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.

Sudden Unexpected Infant Death (SUID), including Sudden Infant Death Syndrome (SIDS), is the most common cause of infant death in the United States, accounting for approximately 3,700 annual deaths of children under 1 year of age. Although the etiology of SIDS is unknown, previous works support the hypothesis that impairment of the neonate autoresuscitation reflex is a common endpoint in SIDS. The autoresuscitation reflex is a protective response to hypercapnia and hypoxia in which an infant undergoes deep gasping to restart the cardio-respiratory system. However, measurable in vivo physiological biomarkers that can be used as a diagnostic tool to identify SIDS-vulnerable infants currently do not exist. Thus, the goal of our project was to combine modern machine learning with novel mouse modeling and assaying capabilities as groundwork for identifying biomarkers for SIDS. We leveraged a new automated phenotyping platform, Looper, to expose over 800 C57Bl6/J mouse pups to repeated hypercapnic anoxia challenges and recovery periods that continue in a loop until death. These challenges are designed to induce the autoresuscitation reflex, which allows us to study cardiorespiratory behavior before, during, and after the reflex is triggered. We then applied modern machine learning techniques to interrogate the large-scale cardiorespiratory data. We applied several supervised learning methods to model nonlinear relationships between variables derived from breath-wise and challenge-wise statistics of respiration and cardiovascular activity, including ventilatory frequencies, tidal volumes, heart rates, and recovery times. The best-performing predictive model was selected using nested cross-validation. A post-hoc leave-one-covariate-out approach was then applied to the chosen model to assess the statistical significance of each cardiorespiratory variable for predicting the number of challenges survived. The findings from our analysis showed several cardiorespiratory features related to the recovery phase of the first anoxic challenge, baseline acclimation period, and temporal latency to the recovery of breathing and heart rate after a challenge that are significantly correlated with the expected number of challenges survived. Using the novel data-driven insights gleaned from our machine learning analysis, we aim to derive mechanistic insights on the underlying neuropathological and physiological systems that underlie the control of breathing in neonates during stress, to test the effect of neuronal manipulations on the relationships between survivability and cardiorespiratory outcomes, and to hypothesize potential diagnostics and therapeutic interventions for SIDS. FUNDING: NIH/NHLBI R01HLN161142-01 and R01HL130249-07 This abstract was presented at the American Physiology Summit 2025 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.

SIDS/SUID (henceforth referred to as SIDS) is a leading cause of neonate death, taking the lives of ~10 infants every day in the US alone. SIDS is defined as the death of an infant for which no cause of death can be determined after a complete autopsy, death scene investigation, and medical history review, all of which makes SIDS a very diffcult disease to study. Death scene investigations have identified several contributing environmental factors. However, etiological investigation has been complicated by inconsistencies in classification of death by medical examiners or coroners and unreliable resources for law enforcement to complete death scene investigations, which leads to epidemiological studies often lacking rigor or suffcient data. Using data from Child Fatality Review Team (CFRT) of Harris County, the Medical Examiner’s Offce, and the Texas Department of State Health Services, we performed a retrospective analysis of 17-years of SIDS cases in Harris County, Texas. We applied a novel birth rate correction for SIDS rates over the first year of life. Associations were tested using a Tukey HSD and linear regression models with co-linear factors considered. To focus on biologically unexplained SIDS cases, we subsetted our dataset to only include SIDS and Undetermined causes of death. Our preliminary data includes 1,094 cases comprised of ~43% black, ~32% Hispanic, ~20% white, ~3% Asian, and ~0.4% unknown and 57% male infants. Distribution of death across the first 12 months of life shows a peak of death around 2 months of age. We do not see an association with month or season in the raw SIDS data nor when using our novel correction for birth rates. When we apply the standard per 1,000 live births normalization, we see a significant increase in Spring and Winter. We also report race-dependent significant differences in yearly and monthly case numbers, male-to-female ratios, and age at death and additional significant race-dependent associations between SIDS cases and adults diagnosed with asthma, income level, and overcrowding based on Zip Code. This dataset represents one of the most racially and socio-economically diverse in the United States. Our data suggests that standard trends may be driven by majority white datasets and that response to socioeconomic stress varies by race/ ethnicity, which leads to differences in risk between classically described minority groups. NIH: 1F32HL160073-01A1, R01HL130249 44617-S4. BCM McNair Scholar Program, March of Dimes Basil O'Connor Research Award, Parker B. Francis Fellowship, CJ Foundation for SIDS. This is the full abstract presented at the American Physiology Summit 2024 meeting and is only available in HTML format. There are no additional versions or additional content available for this abstract. Physiology was not involved in the peer review process.

Measurement and analysis of cardio-respiratory function in animal models is essential for our understanding of functional and disease mechanisms that pave the way for human therapeutics and treatments. However, physiological phenotyping in conscious animals is often an onerous process requiring significant observer time, attention, and manual effort while being susceptible to bias and variability that ultimately limits the kinds of questions that can be asked and answered. Once data is collected, comprehensive and accurate analysis is challenging as it often requires either the use of expensive commercial software(s) with limited flexibility to meet the investigator’s needs or lab developed scripts that are often arduous and require specialized skills to utilize, resulting in incremental insights. Here we present an overview of our high throughput physiology assessment and analysis pipeline for interrogating disease mechanisms and enabling drug discovery in congenital cardio-respiratory disorders such as the lethal sudden infant death syndrome (SIDS). We first developed a closed loop robotic platform for automated neonate cardio-respiratory experiments (Looper) that enables automated interventional studies (i.e. hypoxic induction or drug application) based on real-time physiological feedback during the experiment. We show that deployment of multiple Looper platforms in parallel enables an order of magnitude increase in data collection for a single observer over a single workday and is highly scalable to achieve yet greater outputs. When combined with well designed genetically engineered or other disease mouse models, large scale genetic, molecular, and cellular mechanistic insights and drug discovery become feasible through physiological screening paradigms. Next, we developed an open-sourced cardio-respiratory assessment software tool (Breathe Easy) that enables facile deep interrogation of terabytes of data in a high throughput fashion while offering extensive customization capabilities to allow for unique case by case analysis by more skilled users. Breathe Easy output offers near publication ready graphical outputs with significance markings and statistical tables for rapid presentation of results through manuscripts and web-publishing. Lastly, we will present our early work leveraging our robotic and software pipeline to build an analytical machine learning framework that can predict cardio-respiratory failure (death) with high temporal specificity (more than 10 min out) in our SIDS like assay and our efforts to utilize the trained algorithms for 1) real-time prediction, feedback, and intervention application in our SIDS like assays on the Looper platform to screen small molecules for novel therapeutics; 2) elucidating unique physiological mechanisms through comparative training in distinct SIDS models; and 3) determining the potential for these machine learning approaches to be used in human polysomnographic recordings to identify at risk infants. R01HL161142, R01HL130249. This is the full abstract presented at the American Physiology Summit 2024 meeting and is only available in HTML format. There are no additional versions or additional content available for this abstract. Physiology was not involved in the peer review process.

Central noradrenergic (NA) neurons are key constituents of the respiratory homeostatic network. NA dysfunction is implicated in several developmental respiratory disorders including Congenital Central Hyperventilation Syndrome (CCHS), Sudden Infant Death Syndrome (SIDS) and Rett Syndrome. The current unchallenged paradigm in the field, supported by multiple studies, is that glutamate co-transmission in subsets of central NA neurons plays a role in breathing control. If true, NA-glutamate co-transmission may also be mechanistically important in respiratory disorders. However, the requirement of NA-derived glutamate in breathing has not been directly tested and the extent of glutamate co-transmission in the central NA system remains uncharacterized. Therefore, we fully characterized the cumulative fate maps and acute adult expression patterns of all three Vesicular Glutamate Transporters ( Slc17a7 (Vglut1), Slc17a6 (Vglut2), and Slc17a8 (Vglut3)) in NA neurons, identifying a novel, dynamic expression pattern for Vglut2 and an undescribed co-expression domain for Vglut3 in the NA system. In contrast to our initial hypothesis that NA derived glutamate is required to breathing, our functional studies showed that loss of Vglut2 throughout the NA system failed to alter breathing or metabolism under room air, hypercapnia, or hypoxia in unrestrained and unanesthetized mice. These data demonstrate that Vglut2-based glutamatergic signaling within the central NA system is not required for normal baseline breathing and hypercapnic, hypoxic chemosensory reflexes. These outcomes challenge the current understanding of central NA neurons in the control of breathing and suggests that glutamate may not be a critical target to understand NA neuron dysfunction in respiratory diseases.

Understanding the relationship between circuit connectivity and function is crucial for uncovering how the brain implements computation. In the mouse primary visual cortex (V1), excitatory neurons with similar response properties are more likely to be synaptically connected, but previous studies have been limited to within V1, leaving much unknown about broader connectivity rules. In this study, we leverage the millimeter-scale MICrONS dataset to analyze synaptic connectivity and functional properties of individual neurons across cortical layers and areas. Our results reveal that neurons with similar responses are preferentially connected both within and across layers and areas - including feedback connections - suggesting the universality of the 'like-to-like' connectivity across the visual hierarchy. Using a validated digital twin model, we separated neuronal tuning into feature (what neurons respond to) and spatial (receptive field location) components. We found that only the feature component predicts fine-scale synaptic connections, beyond what could be explained by the physical proximity of axons and dendrites. We also found a higher-order rule where postsynaptic neuron cohorts downstream of individual presynaptic cells show greater functional similarity than predicted by a pairwise like-to-like rule. Notably, recurrent neural networks (RNNs) trained on a simple classification task develop connectivity patterns mirroring both pairwise and higher-order rules, with magnitude similar to those in the MICrONS data. Lesion studies in these RNNs reveal that disrupting 'like-to-like' connections has a significantly greater impact on performance compared to lesions of random connections. These findings suggest that these connectivity principles may play a functional role in sensory processing and learning, highlighting shared principles between biological and artificial systems.

Comprehensive and accurate analysis of respiratory and metabolic data is crucial to modelling congenital, pathogenic and degenerative diseases converging on autonomic control failure. A lack of tools for high-throughput analysis of respiratory datasets remains a major challenge. We present Breathe Easy, a novel open-source pipeline for processing raw recordings and associated metadata into operative outcomes, publication-worthy graphs and robust statistical analyses including QQ and residual plots for assumption queries and data transformations. This pipeline uses a facile graphical user interface for uploading data files, setting waveform feature thresholds and defining experimental variables. Breathe Easy was validated against manual selection by experts, which represents the current standard in the field. We demonstrate Breathe Easy's utility by examining a 2-year longitudinal study of an Alzheimer's disease mouse model to assess contributions of forebrain pathology in disordered breathing. Whole body plethysmography has become an important experimental outcome measure for a variety of diseases with primary and secondary respiratory indications. Respiratory dysfunction, while not an initial symptom in many of these disorders, often drives disability or death in patient outcomes. Breathe Easy provides an open-source respiratory analysis tool for all respiratory datasets and represents a necessary improvement upon current analytical methods in the field. KEY POINTS: Respiratory dysfunction is a common endpoint for disability and mortality in many disorders throughout life. Whole body plethysmography in rodents represents a high face-value method for measuring respiratory outcomes in rodent models of these diseases and disorders. Analysis of key respiratory variables remains hindered by manual annotation and analysis that leads to low throughput results that often exclude a majority of the recorded data. Here we present a software suite, Breathe Easy, that automates the process of data selection from raw recordings derived from plethysmography experiments and the analysis of these data into operative outcomes and publication-worthy graphs with statistics. We validate Breathe Easy with a terabyte-scale Alzheimer's dataset that examines the effects of forebrain pathology on respiratory function over 2 years of degeneration.

Probabilistic graphical models have become an important unsupervised learning tool for detecting network structures for a variety of problems, including the estimation of functional neuronal connectivity from two-photon calcium imaging data. However, in the context of calcium imaging, technological limitations only allow for partially overlapping layers of neurons in a brain region of interest to be jointly recorded. In this case, graph estimation for the full data requires inference for edge selection when many pairs of neurons have no simultaneous observations. This leads to the graph quilting problem, which seeks to estimate a graph in the presence of block-missingness in the empirical covariance matrix. Solutions for the graph quilting problem have previously been studied for Gaussian graphical models; however, neural activity data from calcium imaging are often non-Gaussian, thereby requiring a more flexible modelling approach. Thus, in our work, we study two approaches for nonparanormal graph quilting based on the Gaussian copula graphical model, namely, a maximum likelihood procedure and a low rank-based framework. We provide theoretical guarantees on edge recovery for the former approach under similar conditions to those previously developed for the Gaussian setting, and we investigate the empirical performance of both methods using simulations as well as real data calcium imaging data. Our approaches yield more scientifically meaningful functional connectivity estimates compared to existing Gaussian graph quilting methods for this calcium imaging data set.

With modern calcium imaging technology, activities of thousands of neurons can be recorded in vivo. These experiments can potentially provide new insights into intrinsic functional neuronal connectivity, defined as contemporaneous correlations between neuronal activities. As a common tool for estimating conditional dependencies in high-dimensional settings, graphical models are a natural choice for estimating functional connectivity networks. However, raw neuronal activity data presents a unique challenge: the relevant information in the data lies in rare extreme value observations that indicate neuronal firing rather than in the observations near the mean. Existing graphical modeling techniques for extreme values rely on binning or thresholding observations which may not be appropriate for calcium imaging data. In this paper we develop a novel class of graphical models, called the Subbotin graphical model, which finds sparse conditional dependency structures with respect to the extreme value observations without requiring data preprocessing. We first derive the form of the Subbotin graphical model and show the conditions under which it is normalizable. We then study the empirical performance of the Subbotin graphical model and compare it to existing extreme value graphical modeling techniques and functional connectivity models from neuroscience through several simulation studies as well as a real-world calcium imaging data example.

Background The functional understanding of genetic interaction networks and cellular mechanisms governing health and disease requires the dissection, and multifaceted study, of discrete cell subtypes in developing and adult animal models. Recombinase-driven expression of transgenic effector alleles represents a significant and powerful approach to delineate cell populations for functional, molecular, and anatomical studies. In addition to single recombinase systems, the expression of two recombinases in distinct, but partially overlapping, populations allows for more defined target expression. Although the application of this method is becoming increasingly popular, its experimental implementation has been broadly restricted to manipulations of a limited set of common alleles that are often commercially produced at great expense, with costs and technical challenges associated with production of intersectional mouse lines hindering customized approaches to many researchers. Here, we present a simplified CRISPR toolkit for rapid, inexpensive, and facile intersectional allele production. Results Briefly, we produced 7 intersectional mouse lines using a dual recombinase system, one mouse line with a single recombinase system, and three embryonic stem (ES) cell lines that are designed to study the way functional, molecular, and anatomical features relate to each other in building circuits that underlie physiology and behavior. As a proof-of-principle, we applied three of these lines to different neuronal populations for anatomical mapping and functional in vivo investigation of respiratory control. We also generated a mouse line with a single recombinase-responsive allele that controls the expression of the calcium sensor Twitch-2B. This mouse line was applied globally to study the effects of follicle-stimulating hormone (FSH) and luteinizing hormone (LH) on calcium release in the ovarian follicle. Conclusions The lines presented here are representative examples of outcomes possible with the successful application of our genetic toolkit for the facile development of diverse, modifiable animal models. This toolkit will allow labs to create single or dual recombinase effector lines easily for any cell population or subpopulation of interest when paired with the appropriate Cre and FLP recombinase mouse lines or viral vectors. We have made our tools and derivative intersectional mouse and ES cell lines openly available for non-commercial use through publicly curated repositories for plasmid DNA, ES cells, and transgenic mouse lines.

As a tool for estimating networks in high dimensions, graphical models are commonly applied to calcium imaging data to estimate functional neuronal connectivity, i.e. relationships between the activities of neurons. However, in many calcium imaging data sets, the full population of neurons is not recorded simultaneously, but instead in partially overlapping blocks. This leads to the Graph Quilting problem, as first introduced by (Vinci et.al. 2019), in which the goal is to infer the structure of the full graph when only subsets of features are jointly observed. In this paper, we study a novel two-step approach to Graph Quilting, which first imputes the complete covariance matrix using low-rank covariance completion techniques before estimating the graph structure. We introduce three approaches to solve this problem: block singular value decomposition, nuclear norm penalization, and non-convex low-rank factorization. While prior works have studied low-rank matrix completion, we address the challenges brought by the block-wise missingness and are the first to investigate the problem in the context of graph learning. We discuss theoretical properties of the two-step procedure, showing graph selection consistency of one proposed approach by proving novel L infinity-norm error bounds for matrix completion with block-missingness. We then investigate the empirical performance of the proposed methods on simulations and on real-world data examples, through which we show the efficacy of these methods for estimating functional connectivity from calcium imaging data.

This Innovative-Practice Full Paper presents the curriculum development and our experiences in offering a client-facing consulting course in data science. Data science education has seen rapid growth over the past decade. To provide students with hands-on opportunities to work with real data, many data science programs have advocated for and implemented experiential learning opportunities throughout the curriculum, which has been shown in a wide variety of literature to have many benefits. Most experiential learning opportunities in STEM programs are provided through capstone and engineering design courses; this is becoming increasingly the case in data science programs as well where several universities have developed data science capstone programs in which students work with clients on the client's real-world data sets. While client-sponsored capstone projects are an exemplar of experiential learning, they may pose major challenges to implement and can be particularly resource-intensive for institutions; this is especially the case in data science where the legalities of data sharing may come with additional hurdles. Because of this, we were motivated to develop a novel client-facing data science consulting course that provides a unique experiential learning scenario to both undergraduate and graduate students while requiring much fewer resources and legalities. In our novel data science consulting course, groups of students work directly with real clients in a consulting clinic setting to provide data science guidance and short-term help with data science challenges. Through this process, students learn about the diversity of real-world problems in data science, how to lead consultations with clients effectively as a team, how to frame data science challenges and research possible solutions, and how to communicate solutions to clients in reports and presentations. We leveraged best practices in consulting courses developed in business school settings to design our course. Additionally, the consulting course serves as a community service initiative whereby researchers, clinicians, non-profit and government workers, and industry professionals benefit from the advice and short-term help provided through consultation. In this paper, we report how our consulting course is set up, how clients from both within and outside the university can seek help at the consulting clinic, and how the structure of the course enables students to have firsthand experience working on many real-world data science problems with clients. Finally, we discuss how student performance is assessed in this course, the lessons learned from offering this course, and recommendations for other data science programs in universities that wish to design similar courses.

Balancing coverage of design process, teaming, and prototyping is a challenge for instructors in their pursuit of creating the perfect engineering design course. Previous studies have demonstrated that teaming and process-based skills can be acquired in a short period of time by applying a training model. Prototyping skills can also be taught but there is a quandary regarding which tools and machines are critical to student success. In this study we evaluated prototypes produced in a first-year team-based engineering design course. Pre-and post-course surveys on prototyping skill and evaluations of end-of-semester prototypes were used to explore which prototyping tools meaningfully contribute to producing functional final prototypes. Several fascinating results have been uncovered through this exploration of student prototyping, including student skills growth and overreporting skill growth, as well as prototyping progress as a critical factor determining design functionality. Our study shows that when considering which prototyping skills to teach in a first-year design course the question is not how many prototyping skills to teach, but how few an instructor can get away with.

Whole body barometric flow through plethysmography is used to study respiratory output in mouse and rat genetic and disease models. Turn‐key commercial and custom systems are both used to gather multiple data streams including respiratory‐pressure waveforms, O2 and CO2 levels, etc. for a reasonable estimation of tidal volume and key metabolic parameters.