Department of Biophysics, UT Southwestern Medical Center, Dallas, Texas.
Department of Biophysics, UT Southwestern Medical Center, Dallas, Texas; Lyda Hill Department of Bioinformatics, UT Southwestern Medical Center, Dallas, Texas.
Biophys J. 2019 Sep 17;117(6):1012-1028. doi: 10.1016/j.bpj.2019.07.050. Epub 2019 Aug 6.
Recent experimental and computational developments have been pushing the limits of live-cell single-molecule imaging, enabling the monitoring of intermolecular interactions in their native environment with high spatiotemporal resolution. However, interactions are captured only for the labeled subset of molecules, which tends to be a small fraction. As a result, it has remained a challenge to calculate molecular interaction kinetics, in particular association rates, from live-cell single-molecule tracking data. To overcome this challenge, we developed a mathematical modeling-based Framework for the Inference of in Situ Interaction Kinetics (FISIK) from single-molecule imaging data with substoichiometric labeling. FISIK consists of (I) devising a mathematical model of molecular movement and interactions, mimicking the biological system and data-acquisition setup, and (II) estimating the unknown model parameters, including molecular association and dissociation rates, by fitting the model to experimental single-molecule data. Due to the stochastic nature of the model and data, we adapted the method of indirect inference for model calibration. We validated FISIK using a series of tests in which we simulated trajectories of diffusing molecules that interact with each other, considering a wide range of model parameters, and including resolution limitations, tracking errors, and mismatches between the model and the biological system it mimics. We found that FISIK has the sensitivity to determine association and dissociation rates, with accuracy and precision depending on the labeled fraction of molecules and the extent of molecule tracking errors. For cases where the labeled fraction is too low (e.g., to afford accurate tracking), combining dynamic but sparse single-molecule imaging data with almost-whole population oligomer distribution data improves FISIK's performance. All in all, FISIK is a promising approach for the derivation of molecular interaction kinetics in their native environment from single-molecule imaging data with substoichiometric labeling.
最近的实验和计算进展推动了活细胞单分子成像的极限，能够以高时空分辨率监测其天然环境中的分子间相互作用。然而，只有被标记的分子亚群被捕获，而这些分子往往只是一小部分。因此，从活细胞单分子跟踪数据计算分子相互作用动力学，特别是缔合速率，仍然是一个挑战。为了克服这一挑战，我们开发了一种基于数学模型的框架，用于从亚化学计量标记的单分子成像数据中推断原位相互作用动力学（FISIK）。FISIK 由以下两部分组成：（I）设计一种分子运动和相互作用的数学模型，模拟生物系统和数据采集设置；（II）通过将模型拟合到实验单分子数据来估计未知模型参数，包括分子缔合和离解速率。由于模型和数据的随机性，我们采用了间接推断方法来校准模型。我们使用一系列测试来验证 FISIK，在这些测试中，我们模拟了相互作用的扩散分子的轨迹，考虑了广泛的模型参数，包括分辨率限制、跟踪误差以及模型与模拟的生物系统之间的不匹配。我们发现 FISIK 具有确定缔合和离解速率的灵敏度，其准确性和精密度取决于被标记的分子分数和分子跟踪误差的程度。对于被标记的分子分数太低的情况（例如，为了进行准确的跟踪），将动态但稀疏的单分子成像数据与几乎完整的聚集体分布数据相结合，可以提高 FISIK 的性能。总之，FISIK 是一种从亚化学计量标记的单分子成像数据中推导出其天然环境中分子相互作用动力学的很有前途的方法。
