Immunotoxicity of ionic liquid [C14mim]BF4 in rats.

Ionic liquid tetrafluoroborated-1-tetradecyl-3-methylimidazole salt ([C14mim]BF4) immunotoxicity was investigated in rats using three exposure groups (12.5, 25, and 50 mg kg-1), one recovery group (50 mg kg-1), and a control group without any treatment. The findings demonstrated that, at low doses, [C14mim]BF4 could raise WBC, NEU, and MID and lysozyme levels as well as spleen T-lymphocyte stimulation index in rats, however at high doses, the aforementioned indices were dramatically lowered. As the dose was raised, the proportion of RBC and PLT in the blood as well as CD4+ and CD8+ in the spleen increased, but the quantity of immunoglobulin IgG, IgA, and IgM in the serum as well as the number of NK cells in the spleen considerably dropped. Even though there were varying degrees of improvement 30 days after ceasing exposure, all these changes were unable to return to normal, and the number of NK cells was further decreased. The study demonstrates that [C14mim]BF4 can damage the specific immunity and non-specific immunity of rats, and cause immune dysfunction.

Prediction of amyloid aggregation rates by machine learning and feature selection.

A novel data-based machine learning algorithm for predicting amyloid aggregation rates is reported in this paper. Based on a highly nonlinear projection from 16 intrinsic features of a protein and 4 extrinsic features of the environment to the protein aggregation rate, a feedforward fully connected neural network (FCN) with one hidden layer is trained on a dataset composed of 21 different kinds of amyloid proteins and tested on 4 rest proteins. FCN shows a much better performance than traditional algorithms, such as multivariable linear regression and support vector regression, with an average accuracy higher than 90%. Furthermore, by the correlation analysis and the principal component analysis, seven key features, folding energy, HP patterns for helix, sheet and helices cross membrane, pH, ionic strength, and protein concentration, are shown to constitute a minimum feature set for characterizing the amyloid aggregation kinetics.

Modeling fibril fragmentation in real-time.

During the application of the mass-action-equation models to the study of amyloid fiber formation, time-consuming numerical calculations constitute a major bottleneck. To conquer this difficulty, here an alternative efficient method is introduced for the fragmentation-only model. It includes two basic steps: (1) simulate close-formed time-evolutionary equations for the number concentration P(t) derived from the moment-closure method; (2) reconstruct the detailed fiber length distribution based on the knowledge of moments obtained in the first step. Compared to direct calculation, our method speeds up the performance by at least 10,000 times (from days to seconds). The accuracy is also satisfactory if suitable functions for the approximate fibril length distribution are taken. Further application to the sonication studies on PI264-b-PFS48 micelles performed by Guerin et al. confirms our method is very promising for the real-time analysis of the experiments on fibril fragmentation.