结合多次DBSCAN和层次聚类算法的膜蛋白单分子定位超分辨图像分割  

Clustering Segmentation for Single⁃Molecule Localization Super⁃Resolution Image of Membrane Protein by Combining Multi⁃Step DBSCAN and Hierarchical Clustering Algorithm

作　　者：杨建宇[1] 胡芬[1] 邢福临 董浩[1] 侯梦迪 李任植[1] 潘雷霆[1,2,3,4] 许京军 Yang Jianyu;Hu Fen;Xing Fulin;Dong Hao;Hou Mengdi;Imshik Lee;Pan Leiting;Xu Jingjun(Key Laboratory of WeakLight Nonlinear Photonics,Ministry of Education,School of Physics,TEDA Institute of Applied Physics,Nankai University,Tianjin 300071,China;Frontiers Science Center for Cell Responses,State Key Laboratory of Medicinal Chemical Biology,College of Life Sciences,Nankai University,Tianjin 300071,China;Shenzhen Research Institute of Nankai University,Shenzhen 518083,Guangdong,China;Collaborative Innovation Center of Extreme Optics Shanxi University,Taiyuan 030006,Shanxi,China)

机构地区：[1]南开大学弱光非线性光子学教育部重点实验室,物理科学学院,泰达应用物理研究院,天津300071 [2]南开大学细胞应答交叉科学中心,药物化学生物学国家重点实验室,生命科学学院,天津300071 [3]南开大学深圳研究院,广东深圳518083 [4]山西大学极端光学协同创新中心,太原山西030006

出　　处：《中国激光》2023年第3期78-85,共8页Chinese Journal of Lasers

基　　金：广东省基础与应用基础研究重大项目(2020B0301030009);国家重点研发计划(2022YFC3400600);国家自然科学基金(11874231,32227802,12174208,31870843);中国博士后科学基金(2020M680032);天津市自然科学基金(20JCYB⁃JC01010);中央高校基本科研业务费(2122021337,2122021405)。

摘　　要：膜蛋白在细胞膜上的时空分布形式决定了其活性状态及功能,在调控细胞生命活动过程中起着重要作用。单分子定位超分辨成像(SMLM)技术为在纳米尺度解析膜蛋白的空间分布提供了可能,但分辨率的极大提升对图像准确聚类分割提出了更高要求。基于密度的空间聚类算法(DBSCAN)是常用的聚类方法之一,但其对于膜蛋白分布不均匀的SMLM超分辨图像的分割效果往往不太理想。本文提出了一种结合多次DBSCAN和层次聚类的混合聚类算法,该算法以DBSCAN方法为分割基础,通过进一步的面积阈值分析和层次聚类,在保持超分辨点簇图像精确聚类识别的前提下,仍能保留每个点簇内的多次定位信号。将该算法应用于模拟数据集和实验数据分割得到的轮廓系数等性能普遍优于传统DBSCAN算法。这种混合聚类方法为膜蛋白SMLM超分辨图像的聚类分割提供了新思路和新方法,有助于更精准地分析膜蛋白在纳米尺度上的空间分布信息。Objective There are a variety of functional proteins localized on the cell membrane that participate in many crucial cellular processes,such as signal transduction and transmembrane transport.The spatiotemporal distribution of specific membrane proteins largely determines their activity states and functions.It is known that the sizes of membrane proteins and the distances between them are both on a nanometer scale.Owing to diffraction limits,traditional optical microscopy cannot provide the spatial distribution of membrane proteins at the singlemolecule level.Therefore,imaging techniques with strong specificity and high resolution are urgently required to reveal the precise spatial distribution of membrane proteins.Nowadays,singlemolecule localization microscopy(SMLM)offers new opportunities to resolve the detailed distribution information of membrane proteins at the nanoscale,while the great improvement in spatial resolution also presents higher demands for accurate clustering segmentation of images.Densitybased spatial clustering of applications with noise clustering(DBSCAN)is one of the most commonly used clustering methods;however,it shows relatively poor performance in clustering segmentation in SMLM images of membrane proteins with heterogeneous density.To address this issue,we propose a novel clustering method using a combination of a multistep DBSCAN and a hierarchical clustering algorithm.This improved clustering method is based on the traditional DBSCAN method,which combines area threshold analysis and hierarchical clustering.Methods In the present work,we improved the traditional DBSCAN method by introducing a variable neighborhood radius and hierarchical clustering to perform precise image clustering segmentation in the original image(Fig.2).First,we inputted a relatively large parameter(ε1,M1)to perform the DBSCAN calculation.Owing to this relatively large parameter,the excessively discrete points in the original image were removed as noise points.Meanwhile,some of the closepoint clusters merged tog

关 键 词：生物光学 单分子定位超分辨成像 超分辨图像分割 膜蛋白 基于密度的空间聚类算法 层次聚类算法 

分 类 号：O436[机械工程—光学工程]