Causal discovery approach with reinforcement learning for risk factors of type II diabetes mellitus.

BACKGROUND: Statistical correlation analysis is currently the most typically used approach for investigating the risk factors of type 2 diabetes mellitus (T2DM). However, this approach does not readily reveal the causal relationships between risk factors and rarely describes the causal relationships visually. RESULTS: Considering the superiority of reinforcement learning in prediction, a causal discovery approach with reinforcement learning for T2DM risk factors is proposed herein. First, a reinforcement learning model is constructed for T2DM risk factors. Second, the process involved in the causal discovery method for T2DM risk factors is detailed. Finally, several experiments are designed based on diabetes datasets and used to verify the proposed approach. CONCLUSIONS: The experimental results show that the proposed approach improves the accuracy of causality mining between T2DM risk factors and provides new evidence to researchers engaged in T2DM prevention and treatment research.

[Lead compound optimization strategy(5) ­ reducing the hERG cardiac toxicity in drug development].

The potassium channel encoded by the human ether-a-go-go related gene(hERG) plays a very important role in the physiological and pathological processes in human. hERG potassium channel determines the outward currents which facilitate the repolarization of the myocardial cells. Some drugs were withdrawn from the market for the serious side effect of long QT interval and arrhythmia due to blockade of hERG channel. The strategies for lead compound optimization are to reduce inhibitory activity of hERG potassium channel and decrease cardiac toxicity. These methods include reduction of lipophilicity and basicity of amines, introduction of hydroxyl and acidic groups, and restricting conformation.

[Analysis of spectral features based on water content of desert vegetation].

By using HR-768 field-portable spectroradiometer made by the Spectra Vista Corporation (SVC) of America, the hyper-spectral data of nine types of desert plants were measured, and the water content of corresponding vegetation was determined by roasting in lab. The continuum of measured hyperspectral data was removed by using ENVI, and the relationship between the water content of vegetation and the reflectance spectrum was analyzed by using correlation coefficient method. The result shows that the correlation between the bands from 978 to 1030 nm and water content of vegetation is weak while it is better for the bands from 1133 to 1266 nm. The bands from 1374 to 1534 nm are the characteristic bands because of the correlation between them and water content is the best. By using cluster analysis and according to the water content, the vegetation could be marked off into three grades: high (>70%), medium (50%-70%) and low (<50%). The research reveals the relationship between water content of desert vegetation and hyperspectral data, and provides basis for the analysis of area in desert and the monitoring of desert vegetation by using remote sensing data.