Graph neural networks (GNN) has been considered as an attractive modelling method for molecular property prediction, and numerous studies have shown that GNN could yield more promising results than traditional descriptor-based methods. In this study, based on 11 public datasets covering various property endpoints, the predictive capacity and computational efficiency of the prediction models developed by eight machine learning (ML) algorithms, including four descriptor-based models (SVM, XGBoost, RF and DNN) and four graph-based models (GCN, GAT, MPNN and Attentive FP), were extensively tested and compared. The results demonstrate that on average the descriptor-based models outperform the graph-based models in terms of prediction accuracy and computational efficiency. SVM generally achieves the best predictions for the regression tasks. Both RF and XGBoost can achieve reliable predictions for the classification tasks, and some of the graph-based models, such as Attentive FP and GCN, can yield outstanding performance for a fraction of larger or multi-task datasets. In terms of computational cost, XGBoost and RF are the two most efficient algorithms and only need a few seconds to train a model even for a large dataset. The model interpretations by the SHAP method can effectively explore the established domain knowledge for the descriptor-based models. Finally, we explored use of these models for virtual screening (VS) towards HIV and demonstrated that different ML algorithms offer diverse VS profiles. All in all, we believe that the off-the-shelf descriptor-based models still can be directly employed to accurately predict various chemical endpoints with excellent computability and interpretability.
For a class of nonlinear least squares problems, it is usually very beneficial to separate the variables into a linear and a nonlinear part and take full advantage of reliable linear least squares techniques. Consequently, the original problem is turned into a reduced problem which involves only nonlinear parameters. We consider in this paper four separated algorithms for such problems. The first one is the variable projection (VP) algorithm with full Jacobian matrix of Golub and Pereyra. The second and third ones are VP algorithms with simplified Jacobian matrices proposed by Kaufman and Ruano et al. respectively. The fourth one only uses the gradient of the reduced problem. Monte Carlo experiments are conducted to compare the performance of these four algorithms. From the results of the experiments, we find that: 1) the simplified Jacobian proposed by Ruano et al. is not a good choice for the VP algorithm; moreover, it may render the algorithm hard to converge; 2) the fourth algorithm perform moderately among these four algorithms; 3) the VP algorithm with the full Jacobian matrix perform more stable than that of the VP algorithm with Kuafman's simplified one; and 4) the combination of VP algorithm and Levenberg-Marquardt method is more effective than the combination of VP algorithm and Gauss-Newton method.
Astragalus membranaceus (AM) has been widely used for treating liver diseases in traditional Chinese medicine. Experimental evidence indicates that it has antitumor potential. In this study, the effect of AM on hepatocarcinogenesis induced by diethylnitrosamine (DEN), two-thirds partial hepatectomy, and 2-acetylaminofluorene (2-AAF) (DEN-PH-AAF) was evaluated using glutathione S-transferase placenta form (GST-P) as marker. First, rats were injected intraperitoneally (i.p.) with DEN (200 mg/kg in saline), a two-thirds partial hepatectomy was carried out 2 weeks later, and the rats were then placed on a basal diet containing 0.02% AAF from week 3 to week 8 to induce hepatocarcinogenesis. The rats were given AM (90 mg/kg or 180 mg/kg body weight) by gavage from week 3 to week 8 (treatment groups). The formation of GST-P-positive foci and the expression of GST-P protein and mRNA caused by DEN-PH-AAF were reduced in the treatment groups, which clearly suggests that AM is effective in delaying DEN-PH-AAF-induced hepatocarcinogenesis.
Recently, preclinical studies have shown that allogeneic adipose-derived stem cells (ASCs), like bone marrow-derived mesenchymal stem cell (BMSCs) have significant clinical benefits in treating cardiovascular diseases, such as ischemic/infarcted heart. In this study, we tested whether ASCs are also immune tolerant, such that they can be used as universal donor cells for myocardial regenerative therapy. The study also focuses on investigating the potential therapeutic effects of human ASCs (hASCs) for myocardial infarction in xenotransplant model, and compares its effects with that of hBMSCs. The in vitro study confirms the superior proliferation potential and viability of hASCs under normoxic and stressed hypoxic conditions compared with hBMSCs. hASCs also show higher potential in adopting cardiomyocyte phenotype. The major findings of the in vivo study are that (1) both hASCs and hBMSCs implanted into immunocompetent rat hearts with acute myocardial infarction survived the extreme environment of xenogeneic mismatch for 6 weeks; (2) both hASCs and hBMSCs showed significant improvement in myocardial pro/anti-inflammatory cytokine levels with no detectable inflammatory reaction, despite the lack of any immunosuppressive therapy; and (3) hASCs contributed to the remarkable improvement in cardiac function and reduced infarction which was significantly better than that of hBMSC and untreated control groups. Thus, our findings suggest the feasibility of using ASCs, instead of BMSCs, as universal donor cells for xenogeneic or allogeneic cell therapy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.