Incorporating temporal information during feature engineering bolsters emulation of spatio-temporal emergence.

MOTIVATION: Emergent biological dynamics derive from the evolution of lower-level spatial and temporal processes. A long-standing challenge for scientists and engineers is identifying simple low-level rules that give rise to complex higher-level dynamics. High-resolution biological data acquisition enables this identification and has evolved at a rapid pace for both experimental and computational approaches. Simultaneously harnessing the resolution and managing the expense of emerging technologies-e.g. live cell imaging, scRNAseq, agent-based models-requires a deeper understanding of how spatial and temporal axes impact biological systems. Effective emulation is a promising solution to manage the expense of increasingly complex high-resolution computational models. In this research, we focus on the emulation of a tumor microenvironment agent-based model to examine the relationship between spatial and temporal environment features, and emergent tumor properties. RESULTS: Despite significant feature engineering, we find limited predictive capacity of tumor properties from initial system representations. However, incorporating temporal information derived from intermediate simulation states dramatically improves the predictive performance of machine learning models. We train a deep-learning emulator on intermediate simulation states and observe promising enhancements over emulators trained solely on initial conditions. Our results underscore the importance of incorporating temporal information in the evaluation of spatio-temporal emergent behavior. Nevertheless, the emulators exhibit inconsistent performance, suggesting that the underlying model characterizes unique cell populations dynamics that are not easily replaced. AVAILABILITY AND IMPLEMENTATION: All source codes for the agent-based model, emulation, and analyses are publicly available at the corresponding DOIs: 10.5281/zenodo.10622155, 10.5281/zenodo.10611675, 10.5281/zenodo.10621244, respectively.

Model design choices impact biological insight: Unpacking the broad landscape of spatial-temporal model development decisions.

Computational models enable scientists to understand observed dynamics, uncover rules underlying behaviors, predict experimental outcomes, and generate new hypotheses. There are countless modeling approaches that can be used to characterize biological systems, further multiplied when accounting for the variety of model design choices. Many studies focus on the impact of model parameters on model output and performance; fewer studies investigate the impact of model design choices on biological insight. Here we demonstrate why model design choices should be deliberate and intentional in context of the specific research system and question. In this study, we analyze agnostic and broadly applicable modeling choices at three levels-system, cell, and environment-within the same agent-based modeling framework to interrogate their impact on temporal, spatial, and single-cell emergent dynamics. We identify key considerations when making these modeling choices, including the (i) differences between qualitative vs. quantitative results driven by choices in system representation, (ii) impact of cell-to-cell variability choices on cell-level and temporal trends, and (iii) relationship between emergent outcomes and choices of nutrient dynamics in the environment. This generalizable investigation can help guide the choices made when developing biological models that aim to characterize spatial-temporal dynamics.

The in silico lab: Improving academic code using lessons from biology.

"Good code" is often regarded as a nebulous, impractical ideal. Common best practices toward improving code quality can be inaccessible to those without a rigorous computer science or software engineering background, contributing to a gap between advancing scientific research and FAIR practices. We seek to equip researchers with the necessary background and context to tackle the challenge of improving code quality in computational biology research using analogies from biology to synthesize why certain best practices are critical for advancing computational research. Improving code quality requires active stewardship; we encourage researchers to deliberately adopt and share practices that ensure reusability, repeatability, and reproducibility.

Mapping CAR T-Cell Design Space Using Agent-Based Models.

Chimeric antigen receptor (CAR) T-cell therapy shows promise for treating liquid cancers and increasingly for solid tumors as well. While potential design strategies exist to address translational challenges, including the lack of unique tumor antigens and the presence of an immunosuppressive tumor microenvironment, testing all possible design choices in vitro and in vivo is prohibitively expensive, time consuming, and laborious. To address this gap, we extended the modeling framework ARCADE (Agent-based Representation of Cells And Dynamic Environments) to include CAR T-cell agents (CAR T-cell ARCADE, or CARCADE). We conducted in silico experiments to investigate how clinically relevant design choices and inherent tumor features-CAR T-cell dose, CD4+:CD8+ CAR T-cell ratio, CAR-antigen affinity, cancer and healthy cell antigen expression-individually and collectively impact treatment outcomes. Our analysis revealed that tuning CAR affinity modulates IL-2 production by balancing CAR T-cell proliferation and effector function. It also identified a novel multi-feature tuned treatment strategy for balancing selectivity and efficacy and provided insights into how spatial effects can impact relative treatment performance in different contexts. CARCADE facilitates deeper biological understanding of treatment design and could ultimately enable identification of promising treatment strategies to accelerate solid tumor CAR T-cell design-build-test cycles.

Modular microenvironment components reproduce vascular dynamics de novo in a multi-scale agent-based model.

Cells do not exist in isolation; they continuously act within and react to their environment. And this environment is not static; it continuously adapts and responds to cells. Here, we investigate how vascular structure and function impact emergent cell population behavior using an agent-based model (ABM). Our ABM enables researchers to "mix and match" cell agents, subcellular modules, and microenvironment components ranging from simple nutrient sources to complex, realistic vascular architectures that accurately capture hemodynamics. We use this ABM to highlight the bilateral relationship between cells and nearby vasculature, demonstrate the effect of vascular structure on environmental heterogeneity, and emphasize the non-linear, non-intuitive relationship between vascular function and the behavior of cell populations over time. Our ABM is well suited to characterizing in vitro and in vivo studies, with applications from basic science to translational synthetic biology and medicine. The model is freely available at https://github.com/bagherilab/ARCADE. A record of this paper's transparent peer review process is included in the supplemental information.

Relationship Between Diabetes, Stress, and Self-Management to Inform Chronic Disease Product Development: Retrospective Cross-Sectional Study.

BACKGROUND: Technology is rapidly advancing our understanding of how people with diabetes mellitus experience stress. OBJECTIVE: The aim of this study was to explore the relationship between stress and sequelae of diabetes mellitus within a unique data set composed of adults enrolled in a digital diabetes management program, Livongo, in order to inform intervention and product development. METHODS: Participants included 3263 adults under age 65 who were diagnosed with diabetes mellitus and had access to Livongo through their employer between June 2015 and August 2018. Data were collected at time of enrollment and 12 months thereafter, which included demographic information, glycemic control, presence of stress, diabetes distress, diabetes empowerment, behavioral health diagnosis, and utilization of behavioral health-related medication and services. Analysis of variance and chi-square tests compared variables across groups that were based on presence of stress and behavioral health diagnosis or utilization. RESULTS: Fifty-five percent of participants (1808/3263) reported stress at the time of at least 1 blood glucose reading. Fifty-two percent of participants (940/1808) also received at least 1 behavioral health diagnosis or intervention. Compared to their peers, participants with stress reported greater diabetes distress, lower diabetes empowerment, greater insulin use, and poorer glycemic control. Participants with stress and a behavioral health diagnosis/utilization additionally had higher body mass index and duration of illness. CONCLUSIONS: Stress among people with diabetes mellitus is associated with reduced emotional and physical health. Digital products that focus on the whole person by offering both diabetes mellitus self-management tools and behavioral health skills and support can help improve disease-specific and psychosocial outcomes.

Agent-Based Models Predict Emergent Behavior of Heterogeneous Cell Populations in Dynamic Microenvironments.

Computational models are most impactful when they explain and characterize biological phenomena that are non-intuitive, unexpected, or difficult to study experimentally. Countless equation-based models have been built for these purposes, but we have yet to realize the extent to which rules-based models offer an intuitive framework that encourages computational and experimental collaboration. We develop ARCADE, a multi-scale agent-based model to interrogate emergent behavior of heterogeneous cell agents within dynamic microenvironments and demonstrate how complexity of intracellular metabolism and signaling modules impacts emergent dynamics. We perform in silico case studies on context, competition, and heterogeneity to demonstrate the utility of our model for gaining computational and experimental insight. Notably, there exist (i) differences in emergent behavior between colony and tissue contexts, (ii) linear, non-linear, and multimodal consequences of parameter variation on competition in simulated co-cultures, and (iii) variable impact of cell and population heterogeneity on emergent outcomes. Our extensible framework is easily modified to explore numerous biological systems, from tumor microenvironments to microbiomes.

Network inference performance complexity: a consequence of topological, experimental and algorithmic determinants.

MOTIVATION: Network inference algorithms aim to uncover key regulatory interactions governing cellular decision-making, disease progression and therapeutic interventions. Having an accurate blueprint of this regulation is essential for understanding and controlling cell behavior. However, the utility and impact of these approaches are limited because the ways in which various factors shape inference outcomes remain largely unknown. RESULTS: We identify and systematically evaluate determinants of performance-including network properties, experimental design choices and data processing-by developing new metrics that quantify confidence across algorithms in comparable terms. We conducted a multifactorial analysis that demonstrates how stimulus target, regulatory kinetics, induction and resolution dynamics, and noise differentially impact widely used algorithms in significant and previously unrecognized ways. The results show how even if high-quality data are paired with high-performing algorithms, inferred models are sometimes susceptible to giving misleading conclusions. Lastly, we validate these findings and the utility of the confidence metrics using realistic in silico gene regulatory networks. This new characterization approach provides a way to more rigorously interpret how algorithms infer regulation from biological datasets. AVAILABILITY AND IMPLEMENTATION: Code is available at http://github.com/bagherilab/networkinference/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Clinician Perceptions Related to the Use of the CBT-I Coach Mobile App.

Objective: Clinicians' perceptions of CBT-I Coach, a patient-facing mobile app for cognitive-behavioral therapy for insomnia (CBT-I), are critical to its adoption and integration into practice. Diffusion of innovations theory emphasizes the influence of perceptions, including the relative advantage to current practice, the compatibility to clinicians' needs, the complexity, the innovation's trialability, and observability. This study intended to evaluate the use and perceptions of CBT-I Coach among Veterans Affairs (VA)-trained CBT-I clinicians. Participants and Methods: Clinicians (N = 108) were surveyed about their use, feedback, and perceptions of CBT-I Coach a year after the app became available. Results: Overall perceptions of CBT-I Coach were favorable. Fifty percent of clinicians reported using CBT-I Coach, with 98% intending to continue use. The app was perceived to increase sleep diary completion and homework compliance. Clinicians viewed the app as providing accessibility to helpful tools and improving patient engagement. Of those not using the app, 83% endorsed intention to use it. Reasons for nonuse were lack of patient access to smart phones, not being aware of the app, not having time to learn it, and inability to directly access app data. Those who reported using CBT-I Coach had more favorable perceptions across all constructs (p < .01 - p < .001), except relative advantage, compared to nonusers. Users perceived it as less complex and more compatible with their practice than nonusers. Conclusions: Continued efforts are needed to increase adoption and enhance use of CBT-I Coach, as well as study if reported benefits can be evidenced more directly.

Smartphone apps for insomnia: examining existing apps' usability and adherence to evidence-based principles for insomnia management.

Insomnia affects up to 22% of the U.S. adult population. The use of mobile health applications (mHealth apps) has been posited as one way to increase access to evidence-based interventions for insomnia, such as cognitive behavioral therapy for insomnia (CBT-I). The purpose of the current study was to summarize the availability of mHealth apps that focus on providing users with the behavioral and/or cognitive skills to manage insomnia, assess their adherence to evidence-based principles, and examine their usability. The terms "insomnia," "insomnia treatment," and "sleep treatment" were used to search the Apple iTunes and Google Play stores in November 2016. Social network query within the authors' professional networks was also conducted. Apps that met inclusion criteria for the study were downloaded and reviewed by the research team for their general characteristics; inclusion of CBT-I skills, strategies, and principles; and aesthetics and usability. Of the 357 apps initially found, 12 met criteria for further review. Overall, the apps were moderately adherent to CBT-I principles, with a mean app score of 1.44 out of 3.00, and moderately usable, with a mean usability score of 3.54 out of 5.00. Few apps currently exist that utilize evidence-based principles to help users practice the behavioral and cognitive skills shown to manage insomnia. Thus, there are exciting opportunities for clinicians, researchers, and mHealth experts to develop effective apps that can help ease the public health burden of insomnia.

Implementation of a Guided, Digital Cognitive Behavioral Program for Anxiety in Primary Care: Preliminary Findings of Engagement and Effectiveness.

BACKGROUND: Implementation of digital behavioral health programs in primary care (PC) can improve access to care for patients in need. INTRODUCTION: This study provides preliminary data on user engagement and anxiety symptom change among patients referred by their PC provider to a guided, mobile cognitive behavioral program, Lantern. MATERIALS AND METHODS: Adults aged 20-65 years with at least mild anxiety (GAD-7 ≥ 5) during routine clinical screening in two PC practices were offered Lantern. The primary outcome was self-reported anxiety collected at baseline and 2 months. Linear mixed effects modeling was used to examine anxiety symptom reduction from baseline to 2 months. Post hoc analyses evaluated how number of units completed, number of techniques practiced, and days of usage impacted symptom change. RESULTS: Sixty-three participants signed up for Lantern and had both baseline and 2- month GAD-7. A mixed effects model adjusted for age, gender, medical complexity score, and physical health found a significant effect of time on GAD-7 (ß = -2.08, standard error = 0.77, t(62) = -2.71, p = 0.009). Post hoc analyses indicated that mean number of units, techniques, and usage days did not significantly impact GAD-7 change over 2 months. However, there was significantly greater improvement in anxiety in participants who completed at least three techniques. DISCUSSION: Results benchmark to previous studies that have found statistically significant symptom change among participants after 4-9 weeks of face-to-face or Internet-based cognitive behavioral therapy (CBT). CONCLUSIONS: This study suggests that use of Lantern is associated with anxiety reduction and provides proof-of-concept for the dissemination and implementation of guided, CBT-based mobile behavioral health interventions in PC settings.

Modulating and evaluating receptor promiscuity through directed evolution and modeling.

The promiscuity of G-protein-coupled receptors (GPCRs) has broad implications in disease, pharmacology and biosensing. Promiscuity is a particularly crucial consideration for protein engineering, where the ability to modulate and model promiscuity is essential for developing desirable proteins. Here, we present methodologies for (i) modifying GPCR promiscuity using directed evolution and (ii) predicting receptor response and identifying important peptide features using quantitative structure-activity relationship models and grouping-exhaustive feature selection. We apply these methodologies to the yeast pheromone receptor Ste2 and its native ligand α-factor. Using directed evolution, we created Ste2 mutants with altered specificity toward a library of α-factor variants. We then used the  Vectors of Hydrophobic, Steric, and Electronic properties and partial least squares regression to characterize receptor-ligand interactions, identify important ligand positions and properties, and predict receptor response to novel ligands. Together, directed evolution and computational analysis enable the control and evaluation of GPCR promiscuity. These approaches should be broadly useful for the study and engineering of GPCRs and other protein-small molecule interactions.

CellSort: a support vector machine tool for optimizing fluorescence-activated cell sorting and reducing experimental effort.

Motivation: High throughput screening by fluorescence activated cell sorting (FACS) is a common task in protein engineering and directed evolution. It can also be a rate-limiting step if high false positive or negative rates necessitate multiple rounds of enrichment. Current FACS software requires the user to define sorting gates by intuition and is practically limited to two dimensions. In cases when multiple rounds of enrichment are required, the software cannot forecast the enrichment effort required. Results: We have developed CellSort, a support vector machine (SVM) algorithm that identifies optimal sorting gates based on machine learning using positive and negative control populations. CellSort can take advantage of more than two dimensions to enhance the ability to distinguish between populations. We also present a Bayesian approach to predict the number of sorting rounds required to enrich a population from a given library size. This Bayesian approach allowed us to determine strategies for biasing the sorting gates in order to reduce the required number of enrichment rounds. This algorithm should be generally useful for improve sorting outcomes and reducing effort when using FACS. Availability and Implementation: Source code available at http://tyolab.northwestern.edu/tools/ . k-tyo@northwestern.edu. Supplementary information: Supplementary data are available at Bioinformatics online.

Multi-class and multi-scale models of complex biological phenomena.

Computational modeling has significantly impacted our ability to analyze vast (and exponentially increasing) quantities of experimental data for a variety of applications, such as drug discovery and disease forecasting. Single-scale, single-class models persist as the most common group of models, but biological complexity often demands more sophisticated approaches. This review surveys modeling approaches that are multi-class (incorporating multiple model types) and/or multi-scale (accounting for multiple spatial or temporal scales) and describes how these models, and combinations thereof, should be used within the context of the problem statement. We end by highlighting agent-based models as an intuitive, modular, and flexible framework within which multi-scale and multi-class models can be implemented.

Binge/purge thoughts in nonsuicidal self-injurious adolescents: an ecological momentary analysis.

OBJECTIVE: Adolescents who self-injure often engage in bingeing/purging (BP). Ecological momentary assessment (EMA) has potential to offer insight into the relationship between self-injury and BP. The aims of this study were to examine the frequency and context of BP using EMA in a sample of nonsuicidal self-injurious (NSSI) adolescents. METHOD: Thirty adolescents with a history of NSSI responded to questions regarding self-destructive thoughts/behaviors using a palm-pilot device. Descriptive analyses compared thought/behavior contexts during reports of BP and NSSI thoughts/behaviors (occurring together vs. individually). RESULTS: BP thoughts were present in 22 (73%) participants, occurring on 32% of the person-days recorded; 59% of these participants actually engaged in BP behavior. Seventy-nine percent of BP thoughts co-occurred with other self-destructive thoughts. Adolescents were more often with friends/peers than alone or with family when having BP thoughts. Worry and pressure precipitated both BP and NSSI thoughts, but perceived criticism and feelings of rejection/hurt were associated more often with BP thoughts than with NSSI thoughts. DISCUSSION: BP thoughts and behaviors were common in this sample, often occurring with other self-destructive thoughts. Future EMA research is needed to address the function of BP symptoms, the contextual variables that increase risk for BP thoughts, and the factors that predict the transition of thoughts into behaviors in adolescents with and without self-injury.