A Study on Prediction of the Bio‐toxicity of Substituted Benzene Based on Artificial Neural Network

Abstract

Quantitative Structure‐Activity Relationship (QSAR) between the bio‐toxicity of seventy‐eight kinds of substituted benzene chemicals to yeast Saccharomyces cerevisiae (lg(1/C_miz)) and the components of vertex degree autocorrelation vectors (values of A, B, C and D) was studied by using the software of Artificial Neural Network (ANN). The key factors of the autocorrelation descriptors for the lg (1/C_miz) value of yeast Saccharomyces cerevisiae, A [0], A [1], C [3], C [5] and D [3] were selected from twenty‐four descriptors, and were explained theoretically in this paper. The QSAR‐ANN model has been used to predict the bio‐toxicity of twenty‐three substituted benzene chemicals. The correlation between C_miz and LC₅₀ was also discussed, and the liner correlation equation between them was established.

Keywords:

• Substituted benzene
• Yeast Saccharomyces cerevisiae
• Bio‐toxicity
• Autocorrelation descriptor
• Quantitative structure‐activity relationship

Acknowledgment

The authors would like to thank the Harbin Institute of Technology for the financial support of this study.