Title:Using Reduced Amino Acid Alphabet and Biological Properties to Analyze and Predict Animal Neurotoxin Protein
Volume: 21
Issue: 10
Author(s): Yao Yu, Shiyuan Wang, Yakun Wang, Yiyin Cao, Chunlu Yu, Yi Pan, Dongqing Su, Qianzi Lu, Yongchun Zuo*Lei Yang*
Affiliation:
- The State key Laboratory of Reproductive Regulation and Breeding of Grassland Livestock, College of Life Sciences, Inner Mongolia University, Hohhot 010070,China
- College of Bioinformatics Science and Technology, Harbin Medical University, Harbin 150081,China
Keywords:
Neurotoxin protein, reduced amino acid alphabet, biological property, support vector machine, non-toxin protein, pharmacological
tools.
Abstract:
Aims: Because of the high affinity of these animal neurotoxin proteins for some special target site, they
were usually used as pharmacological tools and therapeutic agents in medicine to gain deep insights into the function
of the nervous system.
Background and Objective: The animal neurotoxin proteins are one of the most common functional groups among
the animal toxin proteins. Thus, it was very important to characterize and predict the animal neurotoxin proteins.
Methods: In this study, the differences between the animal neurotoxin proteins and non-toxin proteins were analyzed.
Result: Significant differences were found between them. In addition, the support vector machine was proposed to predict
the animal neurotoxin proteins. The predictive results of our classifier achieved the overall accuracy of 96.46%.
Furthermore, the random forest and k-nearest neighbors were applied to predict the animal neurotoxin proteins.
Conclusion: The compared results indicated that the predictive performances of our classifier were better than other
two algorithms.