基于深度学习的冠状病毒刺突蛋白拉曼特征峰的理论研究  

Theoretical Study on Raman Characteristic Peaks of Coronavirus Spike Protein Based on Deep Learning

作　　者：倪爽 温家星 周民杰[1] 黄景林[1] 乐玮[1] 陈果[1] 何智兵[1] 李波[1] 赵松楠[1] 赵宗清[1] 杜凯[1] NI Shuang;WEN Jia-xing;ZHOU Min-jie;HUANG Jing-lin;LE Wei;CHEN Guo;HE Zhi-bing;LI Bo;ZHAO Song-nan;ZHAO Zong-qing;DU Kai(Laser Fusion Research Center,China Academy of Engineering Physics,Mianyang 621900,China)

机构地区：[1]中国工程物理研究院激光聚变研究中心,四川绵阳621900

出　　处：《光谱学与光谱分析》2022年第9期2757-2762,共6页Spectroscopy and Spectral Analysis

基　　金：国家自然科学基金项目(22008225);国防科工局《新冠病毒拉曼指纹峰标定与快速检测技术验证研究》项目资助。

摘　　要：持续一年的新冠疫情对全球的经济造成了巨大破坏,为了有效控制新冠疫情,快速检测新冠病毒(SARS-CoV-2)是一个急需解决的问题。新冠病毒的刺突蛋白(spikeprotein)是拉曼光谱技术检测新冠病毒的检测点,构建刺突蛋白拉曼特征峰模型对于发展拉曼检测技术快速检测新冠病毒具有重要作用。基于简化的激子模型,利用深度神经网络技术,构建了刺突蛋白的酰胺Ⅰ、Ⅲ特征峰模型,并结合已知可以感染人类的七种冠状病毒(HCoV-229E,HCoV-HKU1,HCoV-NL63,HCoV-OC43,MERS-CoV,SARS-CoV和SARS-CoV-2)刺突蛋白的实验结构,分析了七种冠状病毒刺突蛋白酰胺Ⅰ、Ⅲ特征峰的区别。计算结果表明,七种冠状病毒可以根据毒刺突蛋白的酰胺Ⅰ、Ⅲ特征峰划分为四个组:SARS-CoV-2,SARS-CoV,MERS-CoV形成一个组;HCoV-HKU1,HCoV-NL63形成一个组;HCoV-229E和HCoV-OC43各自独立形成一个组。相同组的冠状病毒刺突蛋白酰胺Ⅰ、Ⅲ峰频率较为接近,通过酰胺Ⅰ、Ⅲ峰的频率较难区分刺突蛋白;不同组的冠状病毒刺突蛋白酰胺Ⅰ、Ⅲ特征峰差异较大,刺突蛋白可以通过拉曼技术区分开来。该结果为发展拉曼检测技术快速检测新冠病毒提供了定性判断的理论依据。COVID-19,which has lasted for a year,has caused great damage to the global economy.In order to control COVID-19 effectively,rapid detection of COVID-19(SARS-CoV-2)is an urgent problem.Spike protein is the detection point of Raman spectroscopy to detect SARS-CoV-2.The construction of spike protein Raman characteristic peaks plays an important role in the rapid detection of SARS-CoV-2 using Raman technology.In this paper,we used Deep Neural Networks to construct the amideⅠandⅢcharacteristic peak model of spike proteins based on simplified exciton model,and combined with the experimental structures of seven coronaviruses(HCoV-229 E,HCoV-HKU1,HCoV-NL63,HCoV-OC43,MERS-CoV,SARS-CoV,SARS-CoV-2)spike proteins,analyzed the differences of amideⅠandⅢcharacteristic peaks of seven coronaviruses.The results showed that seven coronaviruses could be divided into four groups according to the amideⅠandⅢcharacteristic peaks of spike proteins:SARS-CoV-2,SARS-CoV,MERS-CoV form a group;HCoV-HKU1,HCoV-NL63 form a group;HCoV-229 E and HCoV-OC43 form a group independently.The frequency of amideⅠandⅢin the same group is relatively close,and it is difficult to distinguish spike proteins by the frequency of amideⅠandⅢ;the characteristic peaks of amideⅠandⅢin different groups are quite different,and spike proteins can be distinguished by Raman spectroscopy.The results provide a theoretical basis for the development of Raman spectroscopy for rapid detection of SARS-CoV-2.

关 键 词：冠状病毒 刺突蛋白 拉曼光谱 酰胺Ⅰ、Ⅲ峰 深度学习 

分 类 号：O641.12[理学—物理化学]