Are High- or Low-dose SGLT2 Inhibitors Associated With Cardiovascular and Respiratory Adverse Events? A Meta-analysis.

ABSTRACT: The association between high-dose or low-dose sodium-glucose cotransporter 2 (SGLT2) inhibitors and various cardiovascular and respiratory serious adverse events (SAE) is unclear. Our meta-analysis aimed to define the association between high-dose or low-dose SGLT2 inhibitors and 86 kinds of cardiovascular SAE and 58 kinds of respiratory SAE. We included large cardiorenal outcome trials of SGLT2 inhibitors. Meta-analysis was conducted and stratified by the dose of SGLT2 inhibitors (high dose or low dose) to synthesize risk ratio (RR) and 95% confidence interval (CI). We included 9 trials. Compared with placebo, SGLT2 inhibitors used at high dose or low dose were associated with the decreased risks of 6 kinds of cardiovascular SAE [eg, bradycardia (RR, 0.60; 95% CI, 0.41-0.89), atrial fibrillation (RR, 0.79; 95% CI, 0.69-0.92), and hypertensive emergency (RR, 0.34; 95% CI, 0.15-0.78)] and 6 kinds of respiratory SAE [eg, asthma (RR, 0.59; 95% CI, 0.37-0.93), chronic obstructive pulmonary disease (RR 0.77, 95% CI 0.62-0.96), and sleep apnea syndrome (RR 0.37, 95% CI 0.17-0.81)]. SGLT2 inhibitors used at high dose or low dose did not show significant associations with 132 other cardiopulmonary SAE. For any outcome of interest, the subgroup difference according to the dose of SGLT2 inhibitors was not significant (Psubgroup > 0.05). SGLT2 inhibitors used at whether high dose or low dose are associated with the decreased risks of 12 cardiopulmonary disorders (eg, bradycardia, atrial fibrillation, hypertensive emergency, asthma, chronic obstructive pulmonary disease, and sleep apnea syndrome). These findings may suggest the potential efficacy of high- or low-dose SGLT2 inhibitors for the prevention and treatment of these cardiopulmonary disorders.

A multi-source molecular network representation model for protein-protein interactions prediction.

The prediction of potential protein-protein interactions (PPIs) is a critical step in decoding diseases and understanding cellular mechanisms. Traditional biological experiments have identified plenty of potential PPIs in recent years, but this problem is still far from being solved. Hence, there is urgent to develop computational models with good performance and high efficiency to predict potential PPIs. In this study, we propose a multi-source molecular network representation learning model (called MultiPPIs) to predict potential protein-protein interactions. Specifically, we first extract the protein sequence features according to the physicochemical properties of amino acids by utilizing the auto covariance method. Second, a multi-source association network is constructed by integrating the known associations among miRNAs, proteins, lncRNAs, drugs, and diseases. The graph representation learning method, DeepWalk, is adopted to extract the multisource association information of proteins with other biomolecules. In this way, the known protein-protein interaction pairs can be represented as a concatenation of the protein sequence and the multi-source association features of proteins. Finally, the Random Forest classifier and corresponding optimal parameters are used for training and prediction. In the results, MultiPPIs obtains an average 86.03% prediction accuracy with 82.69% sensitivity at the AUC of 93.03% under five-fold cross-validation. The experimental results indicate that MultiPPIs has a good prediction performance and provides valuable insights into the field of potential protein-protein interactions prediction. MultiPPIs is free available at https://github.com/jiboyalab/multiPPIs .