Supplementary data are available at Bioinformatics online.

Chemical acute oral toxicity is an important end point in drug design and environmental risk assessment. However, it is difficult to determine by experiments, and in silico methods are hence developed as an alternative. In this study, a comprehensive data set containing 12, 204 diverse compounds with median lethal dose (LD₅₀) was compiled. These chemicals were classified into four categories, namely categories I, II, III and IV, based on the criterion of the U.S. Environmental Protection Agency (EPA). Then several multiclassification models were developed using five machine learning methods, including support vector machine (SVM), C4.5 decision tree (C4.5), random forest (RF), κ-nearest neighbor (kNN), and naïve Bayes (NB) algorithms, along with MACCS and FP4 fingerprints. One-against-one (OAO) and binary tree (BT) strategies were employed for SVM multiclassification. Performances were measured by two external validation sets containing 1678 and 375 chemicals, separately. The overall accuracy of the MACCS-SVM(OAO) model was 83.0% and 89.9% for external validation sets I and II, respectively, which showed reliable predictive accuracy for each class. In addition, some representative substructures responsible for acute oral toxicity were identified using information gain and substructure frequency analysis methods, which might be very helpful for further study to avoid the toxicity.

No abstract

Invasions of exotic plant species are among the most pervasive and important threats to natural ecosystems. However, the effects of plant invasions on soil processes and soil biota have not been adequately investigated. Changes were studied in soil microbial communities where Mikania micrantha was invading a native forest community in Neilingding Island, Shenzhen, China. The soil microbial community structure (assessed by phospholipid fatty acid [PLFA] profiles) and function (assessed by enzyme activities), as well as soil chemical properties were measured. The results showed that the invasion of M. micrantha into the evergreen broadleaved forests in South China changed most of the characteristics in studied soils. Microbial community structure and function differed significantly among the native, two ecotones, and exotic-derived soils. For PLFA profiles, we observed a significant increase in aerobic bacteria but a decrease in anaerobic bacteria in the M. micrantha monoculture as compared to the native and ecotones. The ratio of cy19:0 to18:1x7 gradually declined but mono/sat PLFAs increased as M. micrantha became more dominant. Both ratios were significantly related to pH according to regression analysis, therefore, pH was a sensitive indicator reflecting the invaded soil subsystem succession. The microbial community composition clearly separated the native soil from the invaded soils by principal component analysis (PCA) and discriminant analysis (DA). For enzyme activities, 7 of 9 enzymes (b-glucosidase, invertase, protease, urease, acid phosphatase, alkaline phosphatase, and phenol oxidase) showed the similar trend that the activities were highest in the exotic, intermediate in the two ecotones, and lowest in the native community. In most cases, enzyme activities were influenced by soil chemical properties, especially by pH value and soil organic matter. Differences in the structural variables were well correlated to differences in the functional variables as demonstrated by canonical correlation analysis (CCA). It was concluded that M. micrantha invasion had profound effects on the soil subsystem, which must be taken into account when we try to control its invasions.

No abstract

No abstract

Slippery liquid-infused porous surfaces, emerging bio-inspired surfaces which have attracted widespread research interest over the past few years, have great potential in both corrosion protection and biofouling prevention.

No abstract