Tracking bacteria at high density with FAST, the Feature-Assisted Segmenter/Tracker.
Tracking bacteria at high density with FAST, the Feature-Assisted Segmenter/Tracker.
复制标题
DOI：
10.1371/journal.pcbi.1011524
复制
发表时间：
2023-10
影响因子：
4.3
通讯作者：
中科院分区：
文献类型：
--
作者：
大多数细菌在密集的群落中附着在表面上。虽然新的实验和成像技术开始为这些群体的复杂过程提供一个窗口，但通过时间和空间解决单个细胞的行为仍然是一个主要挑战。尽管已经开发了许多不同的软件解决方案来跟踪微生物，但这些解决方案通常要求用户要么调整大量参数，要么对大量成像数据进行实地分析，以训练深度学习模型——这两种手动过程对于新实验来说都非常耗时。为了克服这些限制，我们开发了FAST，即特征辅助分割/跟踪器，它使用无监督机器学习来优化跟踪，同时保持易用性。我们的方法根植于信息论，在很大程度上消除了用户手动迭代调整参数和对所得细胞轨迹进行定性评估的需要。相反，FAST测量每个细胞的多个可区分的“特征”，然后自动量化每个特征提供的独特信息的数量。然后，我们使用这些测量来确定来自不同特征的数据应该如何组合以最小化跟踪误差。将我们的算法与单独使用细胞位置的naïve方法进行比较，发现FAST产生的跟踪误差减少了4到10倍。FAST的模块化设计将我们新颖的跟踪方法与分割，广泛的数据可视化，血统分配和手动轨迹校正工具相结合。它还具有高度可扩展性，允许用户从图像中提取自定义信息，并将其无缝集成到下游分析中。因此，FAST能够以最小的用户输入实现高吞吐量，数据丰富的分析。它已经发布，可以在Matlab中使用，也可以作为编译的独立应用程序使用，并且可以在https://bit.ly/3vovDHn上获得，以及大量的教程和详细的文档。我们对细菌行为的了解大部分来自于通过空间和时间追踪单个细胞。例如，人们可以通过量化单个微生物对营养源作出反应时的运动来解开驱动趋化性的机制。然而，在感染、工业过程和环境中，细菌通常生活在密集的社区中，在那里它们表现出在单独细胞中观察不到的独特行为，例如激活依赖接触的武器来杀死它们的邻居。在这些密集的集合中跟踪个体在技术上是具有挑战性的，因为很难只使用它们在每帧中的位置来跟踪细胞。在这里，我们提出了一种新的软件工具，称为FAST，它结合了机器学习和信息理论来优化细胞跟踪。我们的方法使用一系列不同的细胞特征，如形状和荧光强度，以便随着时间的推移更好地区分个体，与传统方法相比，将误差减少了10倍。此外，FAST估计跟踪精度如何随时间变化，提醒用户注意诸如失焦帧等潜在问题。通过以最少的用户输入跟踪不同条件下的细胞，FAST提供了一个新的定量平台来研究细菌如何适应群体生活。
Most bacteria live attached to surfaces in densely-packed communities. While new experimental and imaging techniques are beginning to provide a window on the complex processes that play out in these communities, resolving the behaviour of individual cells through time and space remains a major challenge. Although a number of different software solutions have been developed to track microorganisms, these typically require users either to tune a large number of parameters or to groundtruth a large volume of imaging data to train a deep learning model—both manual processes which can be very time consuming for novel experiments. To overcome these limitations, we have developed FAST, the Feature-Assisted Segmenter/Tracker, which uses unsupervised machine learning to optimise tracking while maintaining ease of use. Our approach, rooted in information theory, largely eliminates the need for users to iteratively adjust parameters manually and make qualitative assessments of the resulting cell trajectories. Instead, FAST measures multiple distinguishing ‘features’ for each cell and then autonomously quantifies the amount of unique information each feature provides. We then use these measurements to determine how data from different features should be combined to minimize tracking errors. Comparing our algorithm with a naïve approach that uses cell position alone revealed that FAST produced 4 to 10 fold fewer tracking errors. The modular design of FAST combines our novel tracking method with tools for segmentation, extensive data visualisation, lineage assignment, and manual track correction. It is also highly extensible, allowing users to extract custom information from images and seamlessly integrate it into downstream analyses. FAST therefore enables high-throughput, data-rich analyses with minimal user input. It has been released for use either in Matlab or as a compiled stand-alone application, and is available at https://bit.ly/3vovDHn, along with extensive tutorials and detailed documentation. Much of what we know about bacterial behaviour comes from tracking solitary cells through space and time. For example, one can unpick the mechanisms that drive chemotaxis by quantifying the movement of individual microbes as they respond to a nutrient source. However, in infections, industrial processes and the environment, bacteria usually live in tightly-packed communities where they display unique behaviours not observed in solitary cells, such as the activation of contact-dependent weapons to kill their neighbours. Tracking individuals in these dense assemblages is technically challenging because it is difficult to follow cells using only their position in each frame. Here we present a new software tool called FAST which combines machine learning and information theory to optimise cell tracking. Our approach uses a range of different cell characteristics such as shape and fluorescent intensity to better distinguish individuals over time, reducing errors up to 10-fold compared to traditional approaches. In addition, FAST estimates how tracking accuracy changes over time, alerting users to potential problems such as out of focus frames. By tracking cells in a diverse range of conditions with minimal user input, FAST provides a new quantitative platform to study how bacteria have adapted to live in groups.
登录
查看更多内容
DOI：
10.1083/jcb.201004104
发表时间：
2010-05-31
期刊：
The Journal of cell biology
影响因子：
--
作者：
Linkert M;Rueden CT;Allan C;Burel JM;Moore W;Patterson A;Loranger B;Moore J;Neves C;Macdonald D;Tarkowska A;Sticco C;Hill E;Rossner M;Eliceiri KW;Swedlow JR
通讯作者：
Swedlow JR
DOI：
10.1073/pnas.1105073108
发表时间：
2011-08-02
影响因子：
11.1
作者：
Jin, Fan;Conrad, Jacinta C.;Wong, Gerard C. L.
通讯作者：
Wong, Gerard C. L.
DOI：
10.1073/pnas.1811722116
发表时间：
2019-01-29
影响因子：
11.1
作者：
Jeckel, Hannah;Jelli, Eric;Drescher, Knut
通讯作者：
Drescher, Knut
影响因子：
64.5
作者：
Basler M;Ho BT;Mekalanos JJ
通讯作者：
Mekalanos JJ
影响因子：
3.6
作者：
Hartmann, Raimo;van Teeseling, Muriel C. F.;Drescher, Knut
通讯作者：
Drescher, Knut