Abstract: Based on the ideas of suggested vector representation for symbolic sequence and its Fourier transform, the resonance recognition model of higher dimension is presented for protein comparison in this paper. The amino acid sequences, two kinds of protein corresponding to, are numerically encoded, the Fourier transforms of them are computed respectively. So, it can obtain the cross-spectral function for the proteins based on the results of Fourier transform. Using the concept of signal-to-noise ratio, the similarity or difference of the two proteins can be distinguished. Computational results show that the novel method is efficient for protein comparison and it is the extension of Cosic’s resonance recognition model.

Key words: protein comparison, numerical representation of higher dimension, Fourier transform of vector sequence, cross-spectral function, signal to noise ratio, resonance recognition model of higher dimension

摘要： 在提出的符号序列的高维数字表达以及高维傅里叶变换概念的基础上，提出了蛋白质比较的新方法——高维共鸣识别。将两种蛋白质对应的氨基酸序列转化为向量序列，分别计算它们对应的向量序列的离散傅里叶变换。据此，定义两个蛋白质序列所对应的交叉谱函数，考查交叉谱函数的信噪比，判断两种蛋白质序列的相似性或差异性。计算结果显示它是蛋白质比对的又一个有效方法，是Cosic一维共鸣识别的拓展。

关键词: 蛋白质比较, 高维数字表达, 向量序列的傅里叶变换, 交叉谱函数, 信噪比, 高维共鸣识别