Search

The lack of publicly available, large, and unbiased datasets is a key bottleneck for the application of machine learning (ML) methods in synthetic chemistry. Data from electronic laboratory notebooks (ELNs) could provide less biased, large datasets, but no such datasets have been made publicly available. The first real-world dataset from the ELNs of a large pharmaceutical company is disclosed and its relationship to high-throughput experimentation (HTE) datasets is described. For chemical yield predictions, a key task in chemical synthesis, an attributed graph neural network (AGNN) performs as good or better than the best previous models on two HTE datasets for the Suzuki and Buchwald-Hartwig reactions. However, training of the AGNN on the ELN dataset does not lead to a predictive model. The implications of using ELN data for training ML-based models are discussed in the context of yield predictions.

The findings of previous research on the association between proton pump inhibitor (PPI) use and the treatment and prevention of coronavirus disease 2019 (COVID-19) are inconsistent. Therefore, this meta-analysis was conducted to clarify the outcomes of patients taking PPIs. This analysis included 14 articles with more than 268,683 subjects. PPI use was not associated with increased or decreased risk of COVID-19 infection (odds ratio [OR] 1.64, 95% confidence interval [CI] = 0.54-5.00, P = 0.39) or mortality (OR = 1.91, 95% CI = 0.86-4.24, P = 0.11). However, PPI use increased the risks of severe disease (OR 1.67, 95% CI = 1.37-2.02, P0.00001) and secondary infection (OR 4.62, 95% CI = 2.55-8.39, P0.00001). In summary, PPI use was not associated with an increased risk of infection and mortality in COVID-19 but appeared to be associated with an increased risk of progression to severe disease and secondary infection. However, more original studies are urgently needed to further clarify the relationship between PPI use and COVID-19.

Previous research on the association between proton pump inhibitor (PPI) use and the risk of progression to high-grade dysplasia (HGD)/esophageal adenocarcinoma (EAC) in Barrett’s Esophagus (BE) patients has generated inconsistent findings. This meta-analysis was performed to clarify the association. We performed a comprehensive search strategy to select relevant studies up to September 2020. Heterogeneity was assessed using the I-squared statistic. Odds ratios (OR) and 95% confidence intervals (CI) were calculated through either fixed-effects or random-effects model. Duration-response was also performed to assess the gain effects of different PPI intake duration. Sensitivity analysis, subgroup analyses, and tests for publication bias or other small-study effects were conducted. Twelve studies with 155,769 subjects were included. The PPI use was associated with the reduced risk of BE progression to HGD/EAC (OR = 0.47, 95% CI = 0.32–0.71). In the duration–response analysis, the estimated OR for decreased risk of HGD/EAC with PPI intake duration of 12 months was 0.81 (95% CI = 0.71–0.91). Sensitivity analysis suggested the results of this meta-analysis were stable. No publication bias was detected. PPI use is associated with a decreased risk of HGD/EAC in patients with BE. For further investigation, that more well-designed studies are still needed to elucidate the protective effect of PPI usage on BE patients to prevent HGD/EAC.

Background Famotidine was reported to potentially provide benefits to Coronavirus Disease 2019 (COVID-19) patients. However, it remains controversial whether it is effective in treating COVID-19. Aims This study aimed to explore whether famotidine use is associated with reduced risk of the severity, death, and intubation for COVID-19 patients. Methods This study was registered on International Prospective Register of Systematic Reviews (PROSPERO) (ID: CRD42020213536). A comprehensive search was performed to identify relevant studies up to October 2020. I-squared statistic and Q-test were utilized to assess the heterogeneity. Pooled risk ratios (RR) and 95% confidence intervals (CI) were calculated through the random effects or fixed effects model according to the heterogeneity. Subgroup analyses, sensitivity analysis, and publication bias assessment were also conducted. Results Five studies including 36,635 subjects were included. We found that famotidine use was associated with a statistically non-significant reduced risk of progression to severe disease, death, and intubation for Coronavirus Disease 2019 (COVID-19) patients (pooled RR was 0.82, 95% CI = 0.52–1.30, P = 0.40). Conclusion Famotidine has no significant protective effect in reducing the risk of developing serious illness, death, and intubation for COVID-19 patients. More original studies are needed to further clarify whether it is associated with reduced risk of the severity, death, and intubation for COVID-19 patients.

There have been reports of potential negative cardiovascular effects from the COVID-19 vaccine, such as myocarditis or pericarditis. This study sought to ascertain the risk of myocarditis/pericarditis after COVID-19 vaccination by conducting an extensive meta-analysis of published cases.A systematic literature search was conducted in 7 online databases by March 31, 2022. Heterogeneity was tested by IA total of 11 studies with 58,620,611 subjects were included. COVID-19 vaccination correlated with an increased risk of myocarditis or pericarditis (RR=2.04; 95% CI=1.33, 3.14). In addition, an increased risk of myocarditis or pericarditis in people who received the second dose of COVID-19 vaccine compared with that in those who received only the first dose of COVID-19 vaccine was also found (RR=4.06; 95% CI=2.08, 7.92). An increased incidence of pericarditis or myocarditis was noted predominantly in those who received BNT162b2 and mRNA-1273 vaccines (RR=2.19; 95% CI=1.46, 3.29 and RR=4.15; 95% CI=1.87, 9.22, respectively).Study results indicate that a higher incidence of myocarditis or pericarditis was found after COVID-19 vaccination. In addition, the risk of developing myocarditis or pericarditis was greater after the second dose than after the first dose. Nevertheless, the risks of myocarditis and pericarditis in COVID-19 vaccine recipients are still significantly lower than the health risks observed in patients with COVID-19. Therefore, the benefits and harms must be carefully assessed to determine the best management option for patients who are in the high-risk group of myocarditis or pericarditis.

Background and Aims During the current Coronavirus Disease 2019 (COVID-19) pandemic, patients with diabetes face disproportionately more. This study was performed to clarify anti-inflammatory effects of anti-diabetic agents on COVID-19 in patients with diabetes. Methods and Results Relevant literature was searched on 15 databases up to November 14, 2020 and was updated on April 13, 2021. The pooled ORs along with 95% CIs were calculated to evaluate combined effects. 31 studies with 66,914 patients were included in qualitative and quantitative synthesis. Meta-analysis showed that metformin was associated with a statistically significant lower mortality (pooled OR = 0.62, 95% CI, 0.50–0.76, p = 0.000) and poor composite outcomes (pooled OR = 0.83, 95% CI, 0.71–0.97, p = 0.022) in diabetic patients with COVID-19. Significance of slight lower mortality remained in sulfonylurea/glinides (pooled OR = 0.93, 95% CI, 0.89–0.98, p = 0.004), but of poor composite outcomes was not (pooled OR = 1.48, 95% CI, 0.61–3.60, p = 0.384). Dipeptidyl peptidase-4 inhibitors (DPP-4 inhibitors) were associated with statistically non-significant lower mortality (pooled OR = 0.95, 95% CI, 0.72–1.26, p = 0.739) or poor composite outcomes (pooled OR = 1.27, 95% CI, 0.91–1.77, p = 0.162) of COVID-19 in diabetic patients. Conclusion Metformin might be beneficial in decreasing mortality and poor composite outcomes in diabetic patients infected with SARS-CoV-2. DPP-4 inhibitors, sulfonylurea/glinides, SGLT-2 inhibitors, and GLP-1RA would not seem to be adverse. There was insufficient evidence to conclude effects of other anti-diabetic agents. Limited by retrospective characteristics, with relative weak capability to verify causality, more prospective studies, especially RCTs are needed. Registration number: PROSPERO-CRD42020221951.

Background: During the current Coronavirus Disease 2019 (COVID-19) pandemic, diabetic patients face disproportionately more. Anti-inflammatory effects of hypoglycemic agents have been reported, and their effects in patients with diabetes and COVID-19 remain controversial. This study was performed to clarify this association. Methods: Relevant literature was searched on China National Knowledge Infrastructure (CNKI), Wanfang Data Knowledge Service Platform, Chinese periodical service platform VIP Database, Sinomed (China Biology Medicine, CBM), MedRxiv, PubMed, ScienceDirect, Web of Science, Ovid Databases (LWW), Springer Link, Wiley Online Library, Oxford Academic, Nature Press Group, Cochrane Library, and BMJ Evidence-Based Medicine up to November 14, 2020. Only observational studies of hypoglycemic agents vs. other drugs or therapy in adult diabetic patients with COVID-19 were included. Data of death and poor composite outcomes were extracted. The pooled effects of hypoglycemic agents on mortality and poor composite outcomes of COVID-19 in diabetic patients, pooled ORs along with 95% CIs, were calculated using the fixed-effects or random-effects models based on heterogeneity assessment. Registration number of PROSPERO is CRD42020221951. Findings: 16 studies were included in the qualitative analysis and 14 studies with 13,371 patients were included in quantitative synthesis. Meta-analysis showed that metformin was associated with a statistically significant lower mortality in diabetic patients with COVID-19 (pooled OR=0·56, 95% CI, 0·39-0·81, p=0·002), especially home use of metformin (pooled OR=0·60, 95%CI, 0·40-0·88, p=0·01). but in-hospital use of metformin was associated with statistically non-significant lower mortality (pooled OR=0·44, 95%CI, 0·15-1·33, p=0·14) or poor composite outcomes (pooled OR=0·93, 95% CI, 0·79-1·11, p=0·44). Dipeptidyl peptidase-4 inhibitors (DPP-4 inhibitors) were associated with statistically non-significant lower mortality (pooled OR=0·63, 95% CI, 0·26-1·56, p=0·32) or poor composite outcomes of COVID-19 in diabetic patients (pooled OR=0·96, 95% CI, 0·74-1·26, p=0·78). Interpretation: Home use of metformin might be beneficial in decreasing mortality in diabetic patients infected with SARS-CoV-2. There is insufficient evidence to conclude that metformin and other hypoglycemic agents are associated with poor composite outcomes. Limited by retrospective characteristics, with relative weak capability to verify causality, more prospective studies, especially RCTs are needed. Funding Statement: This study is financially supported by Hunan Provincial Key Laboratory of Clinical Epidemiology (grant number: 2020ZNDXLCL002). Declaration of Interests: We declare no competing interests.

BackgroundPrevious researches on the association between proton pump inhibitors (PPIs) use and the treatment and prevention of COVID-19 have generated inconsistent findings. Therefore, this Meta-analysis was conducted to clarify the outcome in patients who take PPIs.MethodsWe carried out a systematic search to identify potential studies until November 2020. Heterogeneity was assessed using the I-squared statistic. Odds ratios (ORs) with its 95% confidence intervals (CIs) were calculated by fixed-effects or random-effects models according to the heterogeneity. Sensitivity analyses and tests for publication bias were also performed.ResultsEight articles with more than 268,683 subjects were included. PPI use was not associated with increased or decreased risk of COVID-19 infection (OR:3.16, 95%CI = 0.74-13.43, P=0.12) or mortality risk of COVID-19 patients (OR=1.91, 95% CI=0.86-4.24, P=0.11). While it can add risk of severe disease (OR=1.54, 95% CI=1.20-1.99, PConclusionsPPI use is not associated with increased risk infection and may not change the mortality risk of COVID-19, but appeared to be associated with increased risk of progression to severe disease and secondary infection. However, more original studies to further clarify the relationship between PPI and COVID-19 are still urgently needed.

With the recent advancement of deep learning, molecular representation learning -- automating the discovery of feature representation of molecular structure, has attracted significant attention from both chemists and machine learning researchers. Deep learning can facilitate a variety of downstream applications, including bio-property prediction, chemical reaction prediction, etc. Despite the fact that current SMILES string or molecular graph molecular representation learning algorithms (via sequence modeling and graph neural networks, respectively) have achieved promising results, there is no work to integrate the capabilities of both approaches in preserving molecular characteristics (e.g, atomic cluster, chemical bond) for further improvement. In this paper, we propose GraSeq, a joint graph and sequence representation learning model for molecular property prediction. Specifically, GraSeq makes a complementary combination of graph neural networks and recurrent neural networks for modeling two types of molecular inputs, respectively. In addition, it is trained by the multitask loss of unsupervised reconstruction and various downstream tasks, using limited size of labeled datasets. In a variety of chemical property prediction tests, we demonstrate that our GraSeq model achieves better performance than state-of-the-art approaches.

BackgroundThe effect of using Angiotensin-converting enzyme inhibitors (ACEIs) and Angiotensin-receptor blockers (ARBs) on the risk of coronavirus disease 2019 (COVID-19) is a topic of recent debate. Although studies have examined the potential association between them, the results remain controversial. This study aims to determine the true effect of ACEI/ARBs use on the risk of infection and clinical outcome of COVID-19.MethodsFive electronic databases (PubMed, Web of science, Cochrane library, China National Knowledge Infrastructure database, medRxiv preprint server) were retrieved to find eligible studies. Meta-analysis was performed to examine the association between ACEI/ARBs use and the risk of infection and clinical outcome of COVID-19.Results22 articles containing 157,328 patients were included. Use of ACEI/ARBs was not associated with increased risk of infection (Adjusted OR: 0.96, 95% CI: 0.91-1.01, I2=5.8%) or increased severity (Adjusted OR: 0.90, 95% CI: 0.77-1.05, I2=27.6%) of COVID-19. The use of ACEI/ARBs was associated with lower risk of death from COVID-19 (Adjusted OR: 0.66, 95% CI: 0.44-0.99, I2=57.9%). Similar results of reduced risk of death were also found for ACEI/ARB use in COVID-19 patients with hypertension (Adjusted OR: 0.36, 95% CI: 0.17-0.77, I2=0).ConclusionThis study provides evidence that ACEI/ARBs use for COVID-19 patients does not lead to harmful outcomes and may even provide a beneficial role and decrease mortality from COVID-19. Clinicians should not discontinue ACEI/ARBs for patients diagnosed with COVID-19 if they are already on these agents.

Given the diverse focuses of emerging online social networks (OSNs), it is common that a user has signed up on multiple OSNs. Social hub services, a.k.a., social directory services, help each user manage and exhibit her OSN accounts on one webpage. In this work, we conduct a data-driven study by crawling over one million user profiles from about.me, a representative online social hub service. Our study aims at gaining insights on cross-OSN social influence from the crawled data. We first analyze the composition of the social hub users. For each user, we collect her social accounts from her social hub webpage, and aggregate the content generated by these accounts on different OSNs to gain a comprehensive view of this user. According to our analysis, there is a high probability that a user would provide consistent information on different OSNs. We then explore the correlation between user activities on different OSNs, based on which we propose a cross-OSN social influence prediction model. With the model, we can accurately predict a user’s social influence on emerging OSNs, such as Instagram, Foursquare, and Flickr, based on her data published on well-established OSNs like Twitter.

Introduction Type 2 diabetes mellitus (T2DM) is a prevalent disease with incidences increasing globally at a rapid rate. The goal of T2DM treatment is to control glucose levels and prevent the aggravation of glycemic symptoms. Treatment options T2DM regimen include metformin as the first-line, with sulfonylurea, thiazolidinedione (TZD), GLP-1, DPP4I, and SGLT2 inhibitor as the second-line treatment options. However, even with a multitude of choices, patient-to-patient variability due to pharmacogenomic differences still prevail. Conclusion This review aims to discuss the responses of the major T2DM medications influenced by pharmacogenomics and investigate improved personalized therapy for T2DM patients.