Automatic classification and segmentation of single-molecule fluorescence time traces with deep learning.

Automatic classification and segmentation of single-molecule fluorescence time traces with deep learning.
复制标题

利用深度学习对单分子荧光时间轨迹进行自动分类和分割

DOI:
10.1038/s41467-020-19673-1
复制
发表时间:
2020-11-17
影响因子:
16.6
通讯作者:
Walter NG
Walter NG
中科院分区:
综合性期刊1区
文献类型:
--
作者:
Li J;Zhang L;Johnson-Buck A;Walter NG

文献摘要

参考文献

被引文献

相似文献

单分子荧光显微镜(SMFM)实验的轨迹呈现出光物理伪影,这通常需要人类专家进行筛选,这种筛选既耗时,又可能引入依赖使用者预期的偏差。在此,我们利用深度学习开发了一种快速、自动的SMFM轨迹选择器,名为AutoSiM,它提高了基于平衡泊松采样(SiMREPS)的单分子识别来检测DNA点突变的灵敏度和特异性。AutoSiM性能的提高是基于相较于传统的隐马尔可夫模型(HMM)后接硬阈值的方法,它能接受更多的真阳性和更少的假阳性。作为第二个应用,该选择器用于自动筛选单分子福斯特共振能量转移(smFRET)数据,以识别用于进一步分析的高质量轨迹,并且在所需处理时间更少的情况下,与人工选择达到约90%的一致性。最后,我们表明AutoSiM能够很容易地适应新的数据集,仅需适度的迁移学习。 单分子荧光显微镜(SMFM)实验的轨迹呈现出光物理伪影,这通常使分析耗时。在此,作者开发了一种易于获取的软件AutoSiM,用于深度学习在高效处理SMFM时间轨迹方面的两种不同应用。
Traces from single-molecule fluorescence microscopy (SMFM) experiments exhibit photophysical artifacts that typically necessitate human expert screening, which is time-consuming and introduces potential for user-dependent expectation bias. Here, we use deep learning to develop a rapid, automatic SMFM trace selector, termed AutoSiM, that improves the sensitivity and specificity of an assay for a DNA point mutation based on single-molecule recognition through equilibrium Poisson sampling (SiMREPS). The improved performance of AutoSiM is based on accepting both more true positives and fewer false positives than the conventional approach of hidden Markov modeling (HMM) followed by hard thresholding. As a second application, the selector is used for automated screening of single-molecule Förster resonance energy transfer (smFRET) data to identify high-quality traces for further analysis, and achieves ~90% concordance with manual selection while requiring less processing time. Finally, we show that AutoSiM can be adapted readily to novel datasets, requiring only modest Transfer Learning. Traces from single-molecule fluorescence microscopy (SMFM) experiments exhibit photophysical artifacts that typically make analysis time-consuming. Here, the authors have developed an easily accessible software, AutoSiM, for two distinct applications of deep learning to the efficient processing of SMFM time traces.
DOI: 10.1016/j.ymeth.2018.08.002
发表时间: 2019-01-15
期刊: Methods (San Diego, Calif.)
影响因子: --
作者:
Johnson-Buck A;Li J;Tewari M;Walter NG
通讯作者: Walter NG
DOI: 10.7554/elife.40183
发表时间: 2018-12-13
期刊: ELIFE
影响因子: 7.7
作者:
Laine, Romain F.;Goodfellow, Gemma;Kaminski, Clemens F.
通讯作者: Kaminski, Clemens F.
DOI: 10.1529/biophysj.105.069252
发表时间: 2006-02-01
影响因子: 3.4
作者:
Margeat, E;Kapanidis, AN;Weiss, S
通讯作者: Weiss, S
DOI: 10.1073/pnas.0401690101
发表时间: 2004-06-15
影响因子: 11.1
作者:
Kapanidis, AN;Lee, NK;Weiss, S
通讯作者: Weiss, S
DOI: 10.1006/meth.2001.1217
发表时间: 2001-09-01
期刊: METHODS
影响因子: 4.8
作者:
Ha, T
通讯作者: Ha, T