Creating cell-specific computational models of stem cell-derived cardiomyocytes using optical experiments.

Human induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs) have gained traction as a powerful model in cardiac disease and therapeutics research, since iPSCs are self-renewing and can be derived from healthy and diseased patients without invasive surgery. However, current iPSC-CM differentiation methods produce cardiomyocytes with immature, fetal-like electrophysiological phenotypes, and the variety of maturation protocols in the literature results in phenotypic differences between labs. Heterogeneity of iPSC donor genetic backgrounds contributes to additional phenotypic variability. Several mathematical models of iPSC-CM electrophysiology have been developed to help understand the ionic underpinnings of, and to simulate, various cell responses, but these models individually do not capture the phenotypic variability observed in iPSC-CMs. Here, we tackle these limitations by developing a computational pipeline to calibrate cell preparation-specific iPSC-CM electrophysiological parameters. We used the genetic algorithm (GA), a heuristic parameter calibration method, to tune ion channel parameters in a mathematical model of iPSC-CM physiology. To systematically optimize an experimental protocol that generates sufficient data for parameter calibration, we created simulated datasets by applying various protocols to a population of in silico cells with known conductance variations, and we fitted to those datasets. We found that calibrating models to voltage and calcium transient data under 3 varied experimental conditions, including electrical pacing combined with ion channel blockade and changing buffer ion concentrations, improved model parameter estimates and model predictions of unseen channel block responses. This observation held regardless of whether the fitted data were normalized, suggesting that normalized fluorescence recordings, which are more accessible and higher throughput than patch clamp recordings, could sufficiently inform conductance parameters. Therefore, this computational pipeline can be applied to different iPSC-CM preparations to determine cell line-specific ion channel properties and understand the mechanisms behind variability in perturbation responses.

Using AI and computer vision to analyze technical proficiency in robotic surgery.

BACKGROUND: Intraoperative skills assessment is time-consuming and subjective; an efficient and objective computer vision-based approach for feedback is desired. In this work, we aim to design and validate an interpretable automated method to evaluate technical proficiency using colorectal robotic surgery videos with artificial intelligence. METHODS: 92 curated clips of peritoneal closure were characterized by both board-certified surgeons and a computer vision AI algorithm to compare the measures of surgical skill. For human ratings, six surgeons graded clips according to the GEARS assessment tool; for AI assessment, deep learning computer vision algorithms for surgical tool detection and tracking were developed and implemented. RESULTS: For the GEARS category of efficiency, we observe a positive correlation between human expert ratings of technical efficiency and AI-determined total tool movement (r = - 0.72). Additionally, we show that more proficient surgeons perform closure with significantly less tool movement compared to less proficient surgeons (p < 0.001). For the GEARS category of bimanual dexterity, a positive correlation between expert ratings of bimanual dexterity and the AI model's calculated measure of bimanual movement based on simultaneous tool movement (r = 0.48) was also observed. On average, we also find that higher skill clips have significantly more simultaneous movement in both hands compared to lower skill clips (p < 0.001). CONCLUSIONS: In this study, measurements of technical proficiency extracted from AI algorithms are shown to correlate with those given by expert surgeons. Although we target measurements of efficiency and bimanual dexterity, this work suggests that artificial intelligence through computer vision holds promise for efficiently standardizing grading of surgical technique, which may help in surgical skills training.

Nurturing diversity and inclusion in AI in Biomedicine through a virtual summer program for high school students.

Artificial Intelligence (AI) has the power to improve our lives through a wide variety of applications, many of which fall into the healthcare space; however, a lack of diversity is contributing to limitations in how broadly AI can help people. The UCSF AI4ALL program was established in 2019 to address this issue by targeting high school students from underrepresented backgrounds in AI, giving them a chance to learn about AI with a focus on biomedicine, and promoting diversity and inclusion. In 2020, the UCSF AI4ALL three-week program was held entirely online due to the COVID-19 pandemic. Thus, students participated virtually to gain experience with AI, interact with diverse role models in AI, and learn about advancing health through AI. Specifically, they attended lectures in coding and AI, received an in-depth research experience through hands-on projects exploring COVID-19, and engaged in mentoring and personal development sessions with faculty, researchers, industry professionals, and undergraduate and graduate students, many of whom were women and from underrepresented racial and ethnic backgrounds. At the conclusion of the program, the students presented the results of their research projects at the final symposium. Comparison of pre- and post-program survey responses from students demonstrated that after the program, significantly more students were familiar with how to work with data and to evaluate and apply machine learning algorithms. There were also nominally significant increases in the students' knowing people in AI from historically underrepresented groups, feeling confident in discussing AI, and being aware of careers in AI. We found that we were able to engage young students in AI via our online training program and nurture greater diversity in AI. This work can guide AI training programs aspiring to engage and educate students entirely online, and motivate people in AI to strive towards increasing diversity and inclusion in this field.

KEA3: improved kinase enrichment analysis via data integration.

Phosphoproteomics and proteomics experiments capture a global snapshot of the cellular signaling network, but these methods do not directly measure kinase state. Kinase Enrichment Analysis 3 (KEA3) is a webserver application that infers overrepresentation of upstream kinases whose putative substrates are in a user-inputted list of proteins. KEA3 can be applied to analyze data from phosphoproteomics and proteomics studies to predict the upstream kinases responsible for observed differential phosphorylations. The KEA3 background database contains measured and predicted kinase-substrate interactions (KSI), kinase-protein interactions (KPI), and interactions supported by co-expression and co-occurrence data. To benchmark the performance of KEA3, we examined whether KEA3 can predict the perturbed kinase from single-kinase perturbation followed by gene expression experiments, and phosphoproteomics data collected from kinase-targeting small molecules. We show that integrating KSIs and KPIs across data sources to produce a composite ranking improves the recovery of the expected kinase. The KEA3 webserver is available at https://maayanlab.cloud/kea3.

Recruitment of Jub by α-catenin promotes Yki activity and Drosophila wing growth.

The Hippo signaling network controls organ growth through YAP family transcription factors, including the Drosophila Yorkie protein. YAP activity is responsive to both biochemical and biomechanical cues, with one key input being tension within the F-actin cytoskeleton. Several potential mechanisms for the biomechanical regulation of YAP proteins have been described, including tension-dependent recruitment of Ajuba family proteins, which inhibit kinases that inactivate YAP proteins, to adherens junctions. Here, we investigate the mechanism by which the Drosophila Ajuba family protein Jub is recruited to adherens junctions, and the contribution of this recruitment to the regulation of Yorkie. We identify α-catenin as the mechanotransducer responsible for tension-dependent recruitment of Jub by identifying a region of α-catenin that associates with Jub, and by identifying a region, which when deleted, allows constitutive, tension-independent recruitment of Jub. We also show that increased Jub recruitment to α-catenin is associated with increased Yorkie activity and wing growth, even in the absence of increased cytoskeletal tension. Our observations establish α-catenin as a multi-functional mechanotransducer and confirm Jub recruitment to α-catenin as a key contributor to biomechanical regulation of Hippo signaling.

Multiplex Immuno-Liquid Chromatography-Mass Spectrometry-Parallel Reaction Monitoring (LC-MS-PRM) Quantitation of CD8A, CD4, LAG3, PD1, PD-L1, and PD-L2 in Frozen Human Tissues.

The immune status of tumors critically influences their responsiveness to PD1 blockades and other immune-based therapies. Programmed death ligand 1 (PD-L1) immunohistochemistry (IHC) is a clinically validated predictive biomarker of response to checkpoint-inhibitor therapy in a limited number of clinical settings but is poorly predictive in most. With emerging evidence that multiple pathways and immune-checkpoint proteins may coordinately contribute to the adaptive immune resistance, the identification and quantitation of multiple immune markers in tumor tissue could help identify the controlling pathways in a given patient, guide the selection of optimal therapy, and monitor response to treatment. We developed and validated a sensitive and robust immuno-liquid chromatography-parallel reaction monitoring assay to simultaneously quantify the expression levels of six immune markers (CD8A, CD4, LAG3, PD1, PD-L1, and PD-L2) using as little as 1-2 mg of fresh frozen tissue. The lower limit of quantitation ranged from 0.07 ng/mg protein for PD1 to 1.0 ng/mg protein for CD4. The intrabatch accuracy was within -16.6% to 15.0% for all proteins at all concentrations, and the variation ranged from 0.8% to 14.7%, while interbatch accuracy was within -6.3% to 8.6%, and the variation ranged from 1.3% to 12.8%. The validated assay was then applied to quantify all six biomarkers in different tissues and was confirmed to have sufficient sensitivity (0.07-1.00 ng/mg protein) and reproducibility (variation ranged from 4.3 to 12.0%). In an analysis of 26 cervical tumors, CD8A and CD4 were detected in all tumors, followed by PD-L1 in 85%, LAG-3 in 65%, PD1 in 50%, and PD-L2 in 35%. The strongest correlations were observed between CD8A and CD4 ( r = 0.88) and CD8A and LAG-3 ( r = 0.86). PD1 was not significantly correlated with any of the other proteins tested. This method can be applied to survey the immune signatures across tumor types and tailored to incorporate additional markers as needed.

Epitope Mapping by HDX-MS Elucidates the Surface Coverage of Antigens Associated with High Blocking Efficiency of Antibodies to Birch Pollen Allergen.

Epitopes of a native pollen allergen protein, birch Bet v1, against four of the noncompeting anti-Bet v1 antibodies individually or in combination, were identified by solution-phase amide backbone H/D exchange (HDX) coupled with high-resolution Q-TOF or Orbitrap mass spectrometry. The HDX results indicates that the four anti-Bet v1 antibodies protected specific regions of Bet v1, explaining the difference in their blocking efficiency of each antibody against Bet v1 binding to polyclonal IgEs in Bet v1 allergic patients. An in-house HDX-MS system was further developed to explore the surface protection of Bet v1 in the presence of all four antibodies with 100% sequence coverage and high redundancy. The data demonstrated that four anti-Bet v1 antibodies were able to simultaneously bind to Bet v1 in solution to provide the most effective blocking for 9 of 10 tested IgE donors in an in vitro antibody-blocking assay. For the first time, we have applied HDX to elucidate the therapeutic advantage of combination antibodies compared with individual antibodies in treating Bet v1 induced allergy.

Emergency department utilization patterns and health care needs assessment among parents of Chinese American children in New York City.

A limited number of publications have addressed the health care needs of Chinese American children of recently immigrated parents. We administered a Chinese-language survey to parents presenting to an urban pediatric emergency department (PED) in New York City and at community venues. The survey assessed demographics, access to health care, and utilization/expectations of the PED. Emergency Severity Index scores were recorded for emergency department patients. Three hundred fifteen families (54% in the PED) completed the survey. Of those completed in the PED, 79% sought emergency services because of pediatric referral or because their pediatrician's office was closed. Of our sickest patients with an Emergency Severity Index score of 3 or less, 28% of parents felt that the child was somewhat sick or not sick at all. Although the majority of our Chinese American families utilize the emergency department appropriately, 28% of the parents of our sickest patients did not appreciate the degree of illness of their children.

Comparison of pre- and postcontrast 3D time-of-flight MR angiography for the evaluation of distal intracranial branch occlusions in acute ischemic stroke.

BACKGROUND AND PURPOSE: Three-dimensional time-of-flight (TOF) MR angiography is used routinely in stroke workup to detect arterial occlusions, but a major drawback is its inadequate depiction of vessels with slow or in-plane flow. We hypothesized that the use of contrast-enhanced MR angiography improves delineation of vessels with diminished or absent flow on precontrast MR angiograms. METHODS: Pre- and postcontrast 3D TOF MR angiograms were acquired in 55 consecutive patients with acute stroke. Patency of 480 intracranial vessels was assessed on both the pre- and postcontrast angiograms. Diffusion-weighted (DW) and perfusion-weighted (PW) imaging data were also obtained and results correlated with those of pre- and postcontrast MR angiography. RESULTS: For 50 abnormal vessel segments seen on precontrast MR angiograms, postcontrast MR angiograms resulted in change in the vascular signal intensity in 70% (35 vessel segments); 94% of these changes showed a greater extent of vessel patency. Venous and soft-tissue contrast enhancement had no effect on assessment in 95% of all 480 vessels examined. Interobserver reliability was moderate, with postcontrast interpretation (kappa = 0.48) showing a slight improvement over precontrast interpretation (kappa = 0.41). Good agreement was found between the TOF results and the pooled DW and PW imaging results. CONCLUSIONS: Compared with precontrast 3D TOF MR angiograms, postcontrast 3D TOF angiograms improve assessment of intracranial vessel patency in acutely ischemic vascular territories. In some patients, an improved understanding of acute ischemic stroke was obtained by viewing the pre- and postcontrast images. Postcontrast MR angiography should be included in the MR evaluation of acute stroke.