Objective-To summarize the prevalence of retinal vein occlusion (RVO) from studies in the United States, Europe, Asia, and Australia. Design-Pooled analysis using individual population-based data.Participants-Individual participant data from population-based studies around the world that had ascertained RVO from fundus photographs.Methods-Each study provided data on branch RVO and central RVO by age, sex, and ethnicity. Prevalence rates were directly age and sex standardized to the 2008 world population aged 30 years and older. Estimates were calculated by study and, after pooling, by ethnicity. Summary estimates included studies in which RVO was assessed from fundus photographs on ≥2 fields of both eyes. Main Outcome Measures-Any RVO, CRVO, or BRVO.Results-The combined pooled data contained 68,751 individuals from 15 studies, with participants' ages ranging from 30 to 101 years. In analyses of 11 studies that assessed ≥2 fundus fields of both eyes (n=49,869), the age-and sex-standardized prevalence was 5.20 per 1000 (confidence interval [CI], 4.40-5.99) for any RVO, 4.42 per 1000 (CI, 3.65-5.19) for BRVO, and 0.80 per 1000 (CI, 0.61-0.99) for CRVO. Prevalence varied by race/ethnicity and increased with age, but did not differ by gender. The age-and sex-standardized prevalence of any RVO was 3.7 per 1000 (CI, 2.8-4.6) in whites (5 studies), 3.9 per 1000 (CI, 1.8-6.0) in blacks (1 study), 5.7 per 1000 (CI, 4.5-6.8) in Asians (6 studies), and 6.9 per 1000 (CI,3) in Hispanics (3 studies). Prevalence for CRVO was lower than BRVO in all ethnic populations. On the basis of these data, an estimated (CI, adults are affected by RVO, with 2.5 million (CI, 1.9-3.1) affected by CRVO and 13.9 million (CI, 11.5-16.4) affected by BRVO. Study limitations include non-uniform sampling frames in identifying study participants and in acquisition and grading of RVO data.Conclusions-Our study provides summary data on the prevalence of RVO and suggests that approximately 16 million people may have this condition. Research on preventive and treatment strategies for this sight-threatening eye disease is needed.Retinal vein occlusion (RVO) is one of the most common causes of acquired retinal vascular abnormality in adults and a frequent cause of visual loss. Despite being recognized at least as early as 1855 1 and the subject of more than 3000 publications, there are few data on the prevalence of RVO in the general population, with current estimates derived largely from studies in white populations. [2][3][4] More recently, population data have emerged from other racial/ ethnic groups, such as in Chinese, 5 Hispanics, 6 and Asian Malays. 7 The reported prevalence of RVO varies between 0.3% 4 and 1.6%. 3 The variability in prevalence rates is likely related to the small number of RVO cases in any single study, differing study methodologies (e.g., retinal photography), and possible racial/ethnic differences in distributions of RVO risk factors.Because of these limitations, estimates of RVO prevalence have been imprecise. Furthermo...

While deep learning has revolutionized protein structure prediction, almost all experimentally characterized de novo protein designs have been generated using physically based approaches such as Rosetta. Here we describe a deep learning–based protein sequence design method, ProteinMPNN, with outstanding performance in both in silico and experimental tests. On native protein backbones, ProteinMPNN has a sequence recovery of 52.4%, compared to 32.9% for Rosetta. The amino acid sequence at different positions can be coupled between single or multiple chains, enabling application to a wide range of current protein design challenges. We demonstrate the broad utility and high accuracy of ProteinMPNN using X-ray crystallography, cryoEM and functional studies by rescuing previously failed designs, made using Rosetta or AlphaFold, of protein monomers, cyclic homo-oligomers, tetrahedral nanoparticles, and target binding proteins.

No abstract

Visual acuity generally improved in eyes with BRVO without intervention, although clinically significant improvement beyond 20/40 was uncommon.

While deep learning has revolutionized protein structure prediction, almost all experimentally characterized de novo protein designs have been generated using physically based approaches such as Rosetta. Here we describe a deep learning based protein sequence design method, ProteinMPNN, with outstanding performance in both in silico and experimental tests. The amino acid sequence at different positions can be coupled between single or multiple chains, enabling application to a wide range of current protein design challenges. On native protein backbones, ProteinMPNN has a sequence recovery of 52.4%, compared to 32.9% for Rosetta. Incorporation of noise during training improves sequence recovery on protein structure models, and produces sequences which more robustly encode their structures as assessed using structure prediction algorithms. We demonstrate the broad utility and high accuracy of ProteinMPNN using X-ray crystallography, cryoEM and functional studies by rescuing previously failed designs, made using Rosetta or AlphaFold, of protein monomers, cyclic homo-oligomers, tetrahedral nanoparticles, and target binding proteins.

Deep learning generative approaches provide an opportunity to broadly explore protein structure space beyond the sequences and structures of natural proteins. Here we use deep network hallucination to generate a wide range of symmetric protein homo-oligomers given only a specification of the number of protomers and the protomer length. Crystal structures of 7 designs are very close to the computational models (median RMSD: 0.6 Å), as are 3 cryoEM structures of giant 10 nanometer rings with up to 1550 residues and C33 symmetry; all differ considerably from previously solved structures. Our results highlight the rich diversity of new protein structures that can be generated using deep learning, and pave the way for the design of increasingly complex components for nanomachines and biomaterials.

BackgroundGlucagon-like peptide-1 receptor agonists (GLP-1RAs) act by increasing insulin secretion, decreasing glucagon secretion, slowing gastric emptying, and increasing satiety.ObjectivePublished evidence directly comparing GLP-1RAs with other approved treatments for type 2 diabetes (T2D) was systematically reviewed.MethodsA literature search was performed using MEDLINE and Embase databases to identify papers comparing GLP-1RAs with other classes of glucose-lowering therapy in patients with T2D.ResultsOf the 1303 papers identified, 57 met the prespecified criteria for a high-quality clinical trial or retrospective study. The efficacy and tolerability of approved GLP-1RAs (exenatide twice daily or once weekly, dulaglutide, liraglutide, lixisenatide, and albiglutide) were compared with insulin products (23 prospective studies + seven retrospective studies), dipeptidyl peptidase-4 inhibitors (11 prospective studies + three retrospective studies), sulfonylureas (nine prospective studies + one retrospective study), thiazolidinediones (five prospective studies), and metformin (two prospective studies). GLP-1RAs are effective as a second-line therapy in improving glycemic parameters in patients with T2D. Reductions in glycated hemoglobin from baseline with GLP-1RAs tended to be greater or similar compared with insulin therapy. GLP-1RAs were consistently more effective in reducing body weight than most oral glucose-lowering drugs and insulin and were associated with lower hypoglycemia risk versus insulin or sulfonylureas. GLP-1RAs improved cardiovascular risk factors, and preliminary data suggest they improve cardiovascular outcomes in patients with T2D compared with oral glucose-lowering drugs. However, results from ongoing studies are awaited to confirm these early findings.ConclusionThis systematic review found that GLP-1RAs are an effective class of glucose-lowering drugs for T2D.

The primary limitations of this study are that it included only a limited number of attributes that may not reflect the full complexity of patient choices, diagnosis was self-reported, and patients were recruited from an Internet panel and may not be representative of the T2DM patient population.