Abstract: This paper uses similarity rules, complementary rule and molecular recognition theory to set up a digital coding model of amino acid for investigation into sequence features and their function identification. Cellular automata is used to generate image representation for protein sequences. A protein sequence can be represented by a unique image, and the image considers the interactional actions between amino acids. Many important features hidden in protein sequence can be revealed through its cellular automata image. The protein subcellular location prediction rate reaches 86.4% based on the visualization model.

Key words: protein sequence, visualization, cellular automata

摘要： 利用相似规则、互补规则和分子识别理论建立一种氨基酸数字编码模型用于研究序列特征、功能预测。给出一种新的基于元胞自动机的蛋白质序列图像生成方法，其优点是考虑了氨基酸前后的相互作用，生成的图像与基因序列一一对应，许多隐藏在蛋白质序列中的重要特性通过元胞自动机图可以表现出来。基于蛋白质元胞自动机图所得到的蛋白质伪氨基酸成分，蛋白质亚细胞定位预测成功率可以达到86.4%。

关键词: 蛋白质序列, 可视化, 元胞自动机

CLC Number: 

• TN919.81