QSAR Study on the Safety Evaluation of Compounds

Fucheng Song, Lianhua Cui, Jun Liu, Jiahong Zheng, Fenglin Sun, Yaqiong Luo, Qing Liu, Bing Ge and Hongzong Si

Abstract: The quantitative structure-activity relationship (QSAR) is using the theoretical calculation and statistical analysis software tools to study the chemical structure and biological effects of quantitative relationship. It is based on a traditional structure-activity relationship and the combination of empirical equation mathematical methods commonly used in physical chemistry. Gene expression programming (GEP) is a relatively new evolutionary computation algorithm, which is a kind of good data mining and modeling tools. Heuristic method (HM) has strong ability of function mining and can rapidly find the answer in complex problem. QSAR model can be used to predict the activity of untested compounds or the physical and chemical properties. On the other hand, QSAR can reveal the structural factors which affect the properties of security risk of compounds, and further better understand the mechanism of action of the toxicity.

: Download (free)  CSE-PDF

"Internet Plus" Tea Industry

Guoqiang Hu

Abstract: With the strong development momentum of “Internet Plus”, Internet information technology has gradually become an internal engine to drive China’s economical development. In terms of tea industry, “Internet Plus” plan serves as the external driving force and essential path to promote the development of tea industry. Based on the analysis of the new technologies of Internet (mobile Internet, cloud computing, big data and Internet of Things, etc.), development strategy of “Internet Plus Tea Industry” is put forward in this paper. In addition, this paper elaborates the specific application of “Internet Plus Tea Industry”, and proposes the suggestions and prospects of “Internet Plus Tea Industry” development strategy.

: Download (free)  CSE-PDF

Analysis of Dangerous Chemical Accident Cause based on Apriori Algorithm

Hao Yu

Abstract: In order to identify the association between causes of accidents due to hazardous chemicals, clear and definite the key points of safety supervision, and improve the efficiency of security supervision, this study researches 150 accident investigation report as data source by data mining technology, Apriori algorithm, and finds out the relationship between factors such as time, step, type, severity and cause. By analyzing the 8 valuable rules, it proposes 4 suggestions for improving key safety supervision, thus to promote the optimization of safety supervision strategy of hazardous chemicals.

: Download (free)  CSE-PDF

Quantitative Structure-activity Relationships (QSAR) of a series of 5-HT2C Agonists

Haoyang Sun, Ting Zhao, Xuejing Zhang and Fucheng Song

Abstract: Gene expression programming (GEP) as a novel machine learning technique was firstly used to predict the EC50 of 5-HT2C agonist. Each kind of compound was represented by a series of calculated structural descriptors. GEP model is based on five descriptors which were selected from the descriptors’ pool by heuristic method (HM) to build multivariable linear model. As a remarkable result, the GEP method generated a nonlinear quantitative model with a correlation coefficient and a variance of 0.82 and 0.10 for the training set, 0.78 and 0.18 for the test set, respectively. It indicates that GEP can be applied to study the 5-HT2C agonist, and the predicted results are well consistent with experimental values.

: Download (free)  CSE-PDF