Statins as a risk factor for diabetic retinopathy: a Mendelian randomization and cross-sectional observational study.

BACKGROUND: Diabetic retinopathy (DR) is the foremost cause of vision loss among the global working-age population, and statins are among the most frequently prescribed drugs for lipid management in patients with DR. The exact relationship between statins and DR has not been determined. This study sought to validate the causal association between statins usage and diabetic retinopathy. METHODS: The summary-data-based Mendelian randomization (SMR) method and inverse-variance-weighted Mendelian randomization (IVW-MR) were used to identify the causal relationship between statins and DR via the use of expression quantitative trait loci (eQTL) data for 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) (31,684 blood samples), low density lipoprotein cholesterol-related GWAS data (sample size: 440,546), and DR-related GWAS data (14,584 cases and 176,010 controls). Additionally, a cross-sectional observational study based on the data from the National Health and Nutrition Examination Survey (NHANES) was conducted to supplement the association between DR and statins (sample size: 106,911). The odds ratios (ORs) with corresponding 95% confidence intervals (CIs) was employed to evaluate the results. RESULTS: Based on the results of the MR analysis, HMGCR inhibitors were causally connected with a noticeably greater incidence of DR (IVW: OR = 0.54, 95% CI [0.42, 0.69], p = 0.000002; SMR: OR = 0.66, 95% CI [0.52, 0.84], p = 0.00073). Subgroup analysis revealed that the results were not affected by the severity of DR. The sensitivity analysis revealed the stability and reliability of the MR analysis results. The results from the cross-sectional study based on NHANES also support the association between not taking statins and a decreased risk of DR (OR = 0.54, 95% CI [0.37, 0.79], p = 0.001). CONCLUSIONS: This study revealed that a significant increase in DR risk was causally related to statins use, providing novel insights into the role of statins in DR. However, further investigations are needed to verify these findings.

Anti-VEGF combined with ocular corticosteroids therapy versus anti-VEGF monotherapy for diabetic macular edema focusing on drugs injection times and confounding factors of pseudophakic eyes: A systematic review and meta-analysis.

PURPOSE: To assess the effectiveness and safety of combining intravitreal endothelial growth factor inhibitor (anti-VEGF) and ocular corticosteroids for diabetic macular edema (DME). METHODS: Articles concentrating on the efficacy and safety of combining anti-VEGF and ocular corticosteroids therapy for DME versus anti-VEGF monotherapy was screened systematically. Meta-analysis was conducted on the basis of a protocol registered in the PROSPERO (CRD42023408338) and performed on the extracted continuous variables and dichotomous variables. The outcome was expressed as weighted mean difference (MD) and risk ratio (RR). RESULTS: Add up to 21 studies including 1468 eyes were enrolled in this study. The MD for best-corrected visual acuity (BCVA) improvement at 1/3/6/12-month between the combination therapy group and monotherapy group were 2.56 (95% CI [0.43, 4.70]), 2.46 (95% CI [-0.40, 5.32]), - 1.76 (95% CI [-3.18, -0.34]), - 1.94 (95% CI [-3.87, 0.00]), respectively. The MD for central retinal thickness (CMT) reduction at 1/3/6/12-month between two groups were - 66.27 (95% CI [-101.08, -31.47]), - 33.62 (95% CI [-57.55, -9.70]), - 4.54 (95% CI [-16.84, 7.76]), - 26.67 (95% CI [-41.52, -11.82]), respectively. Additionally, the combination group had higher relative risk of high intraocular pressure and cataract progression events. CONCLUSIONS: Anti-VEGF combined with ocular corticosteroids had a significant advantage over anti-VEGF monotherapy within 3 months of DME treatment, which reached the maximum with increasing anti-VEGF injection times to 3. However, with the prolongation of the treatment cycle, the effect of combined therapy after 6 months was no better than monotherapy, and the side effects of combined therapy were more severe.

GET: a foundation model of transcription across human cell types.

Transcriptional regulation, involving the complex interplay between regulatory sequences and proteins, directs all biological processes. Computational models of transcription lack generalizability to accurately extrapolate in unseen cell types and conditions. Here, we introduce GET, an interpretable foundation model designed to uncover regulatory grammars across 213 human fetal and adult cell types. Relying exclusively on chromatin accessibility data and sequence information, GET achieves experimental-level accuracy in predicting gene expression even in previously unseen cell types. GET showcases remarkable adaptability across new sequencing platforms and assays, enabling regulatory inference across a broad range of cell types and conditions, and uncovering universal and cell type specific transcription factor interaction networks. We evaluated its performance on prediction of regulatory activity, inference of regulatory elements and regulators, and identification of physical interactions between transcription factors. Specifically, we show GET outperforms current models in predicting lentivirus-based massive parallel reporter assay readout with reduced input data. In fetal erythroblasts, we identify distal (>1Mbp) regulatory regions that were missed by previous models. In B cells, we identified a lymphocyte-specific transcription factor-transcription factor interaction that explains the functional significance of a leukemia-risk predisposing germline mutation. In sum, we provide a generalizable and accurate model for transcription together with catalogs of gene regulation and transcription factor interactions, all with cell type specificity.

Harnessing the potential of large language models in medical education: promise and pitfalls.

OBJECTIVES: To provide balanced consideration of the opportunities and challenges associated with integrating Large Language Models (LLMs) throughout the medical school continuum. PROCESS: Narrative review of published literature contextualized by current reports of LLM application in medical education. CONCLUSIONS: LLMs like OpenAI's ChatGPT can potentially revolutionize traditional teaching methodologies. LLMs offer several potential advantages to students, including direct access to vast information, facilitation of personalized learning experiences, and enhancement of clinical skills development. For faculty and instructors, LLMs can facilitate innovative approaches to teaching complex medical concepts and fostering student engagement. Notable challenges of LLMs integration include the risk of fostering academic misconduct, inadvertent overreliance on AI, potential dilution of critical thinking skills, concerns regarding the accuracy and reliability of LLM-generated content, and the possible implications on teaching staff.

Visual Reasoning: From State to Transformation.

Most existing visual reasoning tasks, such as CLEVR in VQA, ignore an important factor, i.e., transformation. They are solely defined to test how well machines understand concepts and relations within static settings, like one image. Such state driven visual reasoning has limitations in reflecting the ability to infer the dynamics between different states, which has shown to be equally important for human cognition in Piaget's theory. To tackle this problem, we propose a novel transformation driven visual reasoning (TVR) task. Given both the initial and final states, the target becomes to infer the corresponding intermediate transformation. Following this definition, a new synthetic dataset namely TRANCE is first constructed on the basis of CLEVR, including three levels of settings, i.e., Basic (single-step transformation), Event (multi-step transformation), and View (multi-step transformation with variant views). Next, we build another real dataset called TRANCO based on COIN, to cover the loss of transformation diversity on TRANCE. Inspired by human reasoning, we propose a three-staged reasoning framework called TranNet, including observing, analyzing, and concluding, to test how recent advanced techniques perform on TVR. Experimental results show that the state-of-the-art visual reasoning models perform well on Basic, but are still far from human-level intelligence on Event, View, and TRANCO. We believe the proposed new paradigm will boost the development of machine visual reasoning. More advanced methods and new problems need to be investigated in this direction.

Pharmacological roles of lncRNAs in diabetic retinopathy with a focus on oxidative stress and inflammation.

Diabetic retinopathy (DR) is a complication caused by abnormal glucose metabolism, which affects the vision and quality of life of patients and severely impacts the society at large.DR has a complex pathogenic process. Evidence from multiple studies have shown that oxidative stress and inflammation play pivotal roles in DR.Additionally, with the rapid development of various genetic detection methods, the abnormal expression of long non-coding RNAs (lncRNAs) have been confirmed to promote the development of DR.Research has demonstrated the potential of lncRNAs as ideal biomarkers and theranostic targets in DR. In this narrative review, we will focus on the research results on mechanisms underlying DR, list lncRNAs confirmed to be closely related to these mechanisms, and discuss their potential clinical application value and limitations.

Artificial intelligence education: An evidence-based medicine approach for consumers, translators, and developers.

Current and future healthcare professionals are generally not trained to cope with the proliferation of artificial intelligence (AI) technology in healthcare. To design a curriculum that caters to variable baseline knowledge and skills, clinicians may be conceptualized as "consumers", "translators", or "developers". The changes required of medical education because of AI innovation are linked to those brought about by evidence-based medicine (EBM). We outline a core curriculum for AI education of future consumers, translators, and developers, emphasizing the links between AI and EBM, with suggestions for how teaching may be integrated into existing curricula. We consider the key barriers to implementation of AI in the medical curriculum: time, resources, variable interest, and knowledge retention. By improving AI literacy rates and fostering a translator- and developer-enriched workforce, innovation may be accelerated for the benefit of patients and practitioners.

Effect of wheat straw pretreated with liquid fraction of digestate from different substrates on anaerobic digestion performance and microbial community characteristics.

The effects of liquid fraction of digestate (LFD) pretreatment on anaerobic digestion (AD) performance and microbial community characteristics were estimated. Prior to AD, LFD (LFDSM, LFDFW, and LFDWS) collected separately from three continuously stirred tank reactors (CSTRs) using swine manure (SM), food waste (FW), and wheat straw (WS) as the mono-substrate was applied to pretreat WS. The results showed that AD with LFD pretreatment resulted in biomethane yields of 240.2-277.9 mL·gVS-1, a 33.57%-54.54% improvement over the yield of the control, and also produced a time saving of 32.26%-46.77%. The pretreatment parameters were optimized for LFD pretreatment. The enhancement effect was in the order of LFDFW > LFDSM > LFDWS. Simultaneously, the cellulose, hemicellulose and lignin contents in the WS and their characteristics (surface properties, crystallinity index, etc.) varied accordingly. The function of the microbial community was strengthened during the pretreatment stage, but the structure of the microbial community had a clear response to the LFD source substrates. Bacteroidetes was the most dominant phyla and was positively correlated with the hydrolysis rate. Consequently, the LFD from the different substrates used as pretreat agents could improve the AD performance of WS.