Multi-Dataset Comparison of Vision Transformers and Convolutional Neural Networks for Detecting Glaucomatous Optic Neuropathy from Fundus Photographs.

Glaucomatous optic neuropathy (GON) can be diagnosed and monitored using fundus photography, a widely available and low-cost approach already adopted for automated screening of ophthalmic diseases such as diabetic retinopathy. Despite this, the lack of validated early screening approaches remains a major obstacle in the prevention of glaucoma-related blindness. Deep learning models have gained significant interest as potential solutions, as these models offer objective and high-throughput methods for processing image-based medical data. While convolutional neural networks (CNN) have been widely utilized for these purposes, more recent advances in the application of Transformer architectures have led to new models, including Vision Transformer (ViT,) that have shown promise in many domains of image analysis. However, previous comparisons of these two architectures have not sufficiently compared models side-by-side with more than a single dataset, making it unclear which model is more generalizable or performs better in different clinical contexts. Our purpose is to investigate comparable ViT and CNN models tasked with GON detection from fundus photos and highlight their respective strengths and weaknesses. We train CNN and ViT models on six unrelated, publicly available databases and compare their performance using well-established statistics including AUC, sensitivity, and specificity. Our results indicate that ViT models often show superior performance when compared with a similarly trained CNN model, particularly when non-glaucomatous images are over-represented in a given dataset. We discuss the clinical implications of these findings and suggest that ViT can further the development of accurate and scalable GON detection for this leading cause of irreversible blindness worldwide.

STAG2 loss rewires oncogenic and developmental programs to promote metastasis in Ewing sarcoma.

The core cohesin subunit STAG2 is recurrently mutated in Ewing sarcoma but its biological role is less clear. Here, we demonstrate that cohesin complexes containing STAG2 occupy enhancer and polycomb repressive complex (PRC2)-marked regulatory regions. Genetic suppression of STAG2 leads to a compensatory increase in cohesin-STAG1 complexes, but not in enhancer-rich regions, and results in reprogramming of cis-chromatin interactions. Strikingly, in STAG2 knockout cells the oncogenic genetic program driven by the fusion transcription factor EWS/FLI1 was highly perturbed, in part due to altered enhancer-promoter contacts. Moreover, loss of STAG2 also disrupted PRC2-mediated regulation of gene expression. Combined, these transcriptional changes converged to modulate EWS/FLI1, migratory, and neurodevelopmental programs. Finally, consistent with clinical observations, functional studies revealed that loss of STAG2 enhances the metastatic potential of Ewing sarcoma xenografts. Our findings demonstrate that STAG2 mutations can alter chromatin architecture and transcriptional programs to promote an aggressive cancer phenotype.

High-throughput Chemical Screening Identifies Focal Adhesion Kinase and Aurora Kinase B Inhibition as a Synergistic Treatment Combination in Ewing Sarcoma.

PURPOSE: Ewing sarcoma is an aggressive solid tumor malignancy of childhood. Although current treatment regimens cure approximately 70% of patients with localized disease, they are ineffective for most patients with metastases or relapse. New treatment combinations are necessary for these patients. EXPERIMENTAL DESIGN: Ewing sarcoma cells are dependent on focal adhesion kinase (FAK) for growth. To identify candidate treatment combinations for Ewing sarcoma, we performed a small-molecule library screen to identify compounds synergistic with FAK inhibitors in impairing Ewing cell growth. The activity of a top-scoring class of compounds was then validated across multiple Ewing cell lines in vitro and in multiple xenograft models of Ewing sarcoma. RESULTS: Numerous Aurora kinase inhibitors scored as synergistic with FAK inhibition in this screen. We found that Aurora kinase B inhibitors were synergistic across a larger range of concentrations than Aurora kinase A inhibitors when combined with FAK inhibitors in multiple Ewing cell lines. The combination of AZD-1152, an Aurora kinase B-selective inhibitor, and PF-562271 or VS-4718, FAK-selective inhibitors, induced apoptosis in Ewing sarcoma cells at concentrations that had minimal effects on survival when cells were treated with either drug alone. We also found that the combination significantly impaired tumor progression in multiple xenograft models of Ewing sarcoma. CONCLUSIONS: FAK and Aurora kinase B inhibitors synergistically impair Ewing sarcoma cell viability and significantly inhibit tumor progression. This study provides preclinical support for the consideration of a clinical trial testing the safety and efficacy of this combination for patients with Ewing sarcoma.

Unmasking host and microbial strategies in the Agrobacterium-plant defense tango.

Coevolutionary forces drive adaptation of both plant-associated microbes and their hosts. Eloquently captured in the Red Queen Hypothesis, the complexity of each plant-pathogen relationship reflects escalating adversarial strategies, but also external biotic and abiotic pressures on both partners. Innate immune responses are triggered by highly conserved pathogen-associated molecular patterns, or PAMPs, that are harbingers of microbial presence. Upon cell surface receptor-mediated recognition of these pathogen-derived molecules, host plants mount a variety of physiological responses to limit pathogen survival and/or invasion. Successful pathogens often rely on secretion systems to translocate host-modulating effectors that subvert plant defenses, thereby increasing virulence. Host plants, in turn, have evolved to recognize these effectors, activating what has typically been characterized as a pathogen-specific form of immunity. Recent data support the notion that PAMP-triggered and effector-triggered defenses are complementary facets of a convergent, albeit differentially regulated, set of immune responses. This review highlights the key players in the plant's recognition and signal transduction pathways, with a focus on the aspects that may limit Agrobacterium tumefaciens infection and the ways it might overcome those defenses. Recent advances in the field include a growing appreciation for the contributions of cytoskeletal dynamics and membrane trafficking to the regulation of these exquisitely tuned defenses. Pathogen counter-defenses frequently manipulate the interwoven hormonal pathways that mediate host responses. Emerging systems-level analyses include host physiological factors such as circadian cycling. The existing literature indicates that varying or even conflicting results from different labs may well be attributable to environmental factors including time of day of infection, temperature, and/or developmental stage of the host plant.

Lysostaphin-functionalized cellulose fibers with antistaphylococcal activity for wound healing applications.

With the emergence of "super bacteria" that are resistant to antibiotics, e.g., methicillin-resistant Staphylococcus aureus, novel antimicrobial therapies are needed to prevent associated hospitalizations and deaths. Bacteriophages and bacteria use cell lytic enzymes to kill host or competing bacteria, respectively, in natural environments. Taking inspiration from nature, we have employed a cell lytic enzyme, lysostaphin (Lst), with specific bactericidal activity against S. aureus, to generate anti-infective bandages. Lst was immobilized onto biocompatible fibers generated by electrospinning homogeneous solutions of cellulose, cellulose-chitosan, and cellulose-poly(methylmethacrylate) (PMMA) from 1-ethyl-3-methylimidazolium acetate ([EMIM][OAc]), room temperature ionic liquid. Electron microscopic analysis shows that these fibers have submicron-scale diameter. The fibers were chemically treated to generate aldehyde groups for the covalent immobilization of Lst. The resulting Lst-functionalized cellulose fibers were processed to obtain bandage preparations that showed activity against S. aureus in an in vitro skin model with low toxicity toward keratinocytes, suggesting good biocompatibility for these materials as antimicrobial matrices in wound healing applications.