Commonsense reasoning aims to empower machines with the human ability to make presumptions about ordinary situations in our daily life. In this paper, we propose a textual inference framework for answering commonsense questions, which effectively utilizes external, structured commonsense knowledge graphs to perform explainable inferences. The framework first grounds a question-answer pair from the semantic space to the knowledge-based symbolic space as a schema graph, a related sub-graph of external knowledge graphs. It represents schema graphs with a novel knowledge-aware graph network module named KA GNE T, and finally scores answers with graph representations. Our model is based on graph convolutional networks and LSTMs, with a hierarchical path-based attention mechanism. The intermediate attention scores make it transparent and interpretable, which thus produce trustworthy inferences. Using ConceptNet as the only external resource for BERT-based models, we achieved state-of-the-art performance on the CommonsenseQA, a large-scale dataset for commonsense reasoning. We open-source our code 1 to the community for future research in knowledge-aware commonsense reasoning.

Background Pseudomonas aeruginosa represents a good model of antibiotic resistance. These organisms have an outer membrane with a low level of permeability to drugs that is often combined with multidrug efflux pumps, enzymatic inactivation of the drug, or alteration of its molecular target. The acute and growing problem of antibiotic resistance of bacteria to conventional antibiotics made it imperative to develop new liposome formulations for antibiotics, and investigate the fusion between liposome and bacterium. Methods In this study, the factors involved in fluid liposome interaction with bacteria have been investigated. We also demonstrated a mechanism of fusion between liposomes (1,2-dipa lmitoyl-sn-glycero-3-phosphocholine [DPPC]/dimyristoylphosphatidylglycerol [DMPG] 9:1, mol/mol) in a fluid state, and intact bacterial cells, by lipid mixing assay. Results The observed fusion process is shown to be mainly dependent on several key factors. Perturbation of liposome fluidity by addition of cholesterol dramatically decreased the degree of fusion with P. aeruginosa from 44% to 5%. It was observed that fusion between fluid liposomes and bacteria and also the bactericidal activities were strongly dependent upon the properties of the bacteria themselves. The level of fusion detected when fluid liposomes were mixed with Escherichia coli (66%) or P. aeruginosa (44%) seems to be correlated to their outer membrane phosphatidylethanolamine (PE) phospholipids composition (91% and 71%, respectively). Divalent cations increased the degree of fusion in the sequence Fe 2+ > Mg 2+ > Ca 2+ > Ba 2+ whereas temperatures lower than the phase transition temperature of DPPC/DMPG (9:1) vesicles decreased their fusion capacity. Acidic as well as basic pHs conferred higher degrees of fusion (54% and 45%, respectively) when compared to neutral pH (35%). Conclusion Based on the results of this study, a possible mechanism involving cationic bridging between bacterial negatively charged lipopolysaccharide and fluid liposomes DMPG phospholipids was outlined. Furthermore, the fluid liposomal-encapsulated tobramycin was prepared, and the in vitro bactericidal effects were also investigated.

No abstract

No abstract