小柯机器人

一种评估SMLM聚类分析算法性能的框架
2023-02-15 10:04

英国伯明翰大学Dylan M. Owen团队近期取得重要工作进展,他们研究开发了一种评估SMLM聚类分析算法性能的框架。相关研究成果2023年2月10日在线发表于《自然—方法学》杂志上。

据介绍,单分子定位显微镜(SMLM)以定位荧光团坐标的形式生成数据。聚类分析是从这些数据中提取具有生物学意义信息的一种有吸引力的方法,并已被广泛应用。尽管有一系列聚类分析算法,但对其性能的评估还没有共识框架。

研究人员使用基于两个度量的系统方法,在模拟实验数据的模拟条件下对聚类算法的成功进行评分。研究人员用七种不同的分析算法演示了该框架:DBSCAN、ToMATo、KDE、FOCAL、CAML、ClusterViSu和SR Tesseler。考虑到最佳表现取决于定位的基本分布,研究人员展示了一种基于统计相似性度量的分析管道,该管道能够选择最合适的算法,并为真实SMLM数据优化分析参数。

总之,研究人员提议,这些标准模拟条件、度量和分析管道可以作为未来分析算法开发和评估的基础。

附:英文原文

Title: A framework for evaluating the performance of SMLM cluster analysis algorithms

Author: Nieves, Daniel J., Pike, Jeremy A., Levet, Florian, Williamson, David J., Baragilly, Mohammed, Oloketuyi, Sandra, de Marco, Ario, Griffi, Juliette, Sage, Daniel, Cohen, Edward A. K., Sibarita, Jean-Baptiste, Heilemann, Mike, Owen, Dylan M.

Issue&Volume: 2023-02-10

Abstract: Single-molecule localization microscopy (SMLM) generates data in the form of coordinates of localized fluorophores. Cluster analysis is an attractive route for extracting biologically meaningful information from such data and has been widely applied. Despite a range of cluster analysis algorithms, there exists no consensus framework for the evaluation of their performance. Here, we use a systematic approach based on two metrics to score the success of clustering algorithms in simulated conditions mimicking experimental data. We demonstrate the framework using seven diverse analysis algorithms: DBSCAN, ToMATo, KDE, FOCAL, CAML, ClusterViSu and SR-Tesseler. Given that the best performer depended on the underlying distribution of localizations, we demonstrate an analysis pipeline based on statistical similarity measures that enables the selection of the most appropriate algorithm, and the optimized analysis parameters for real SMLM data. We propose that these standard simulated conditions, metrics and analysis pipeline become the basis for future analysis algorithm development and evaluation.

DOI: 10.1038/s41592-022-01750-6

Source: https://www.nature.com/articles/s41592-022-01750-6

Nature Methods:《自然—方法学》,创刊于2004年。隶属于施普林格·自然出版集团,最新IF:47.99
官方网址:https://www.nature.com/nmeth/
投稿链接:https://mts-nmeth.nature.com/cgi-bin/main.plex


本期文章:《自然—方法学》:Online/在线发表

分享到:

0