Data for 'Ranking Single Fluorescent Protein Based Calcium Biosensor Performance by Molecular Dynamics Simulations'

Melike Berksoz, Canan Atilgan* Faculty of Engineering and Natural Sciences, Sabanci University *Correspondance: Canan Atilgan, Faculty of Natural Sciences and Engineering, Sabancı University, Tuzla 34956 Istanbul, Türkiye, E-mail: [e-mail] Genetically Encoded Fluorescent Biosensors (GEFBs) have become indispensable tools for visualizing biological processes in vivo. A typical GEFB is composed of a sensory domain (SD) which undergoes a conformational change upon ligand binding and a genetically fused fluorescent protein (FP). Ligand binding in the SD allosterically modulates the chromophore environment and changes its spectral properties. Single fluorescent (FP)-based biosensors, a subclass of GEFBs, offer a simple experimental setup; they are easy to produce in living cells, structurally stable and simple due to their single-wavelength operation. However, they pose a significant challenge for structure optimization, especially concerning the length and residue content of linkers between the FP and SD which effect how well the chromophore responds to conformational change in the SD. In this work, we use classical all-atom molecular dynamics simulations to analyze the dynamic properties of a series of calmodulin-based calcium biosensors, all with different FP-SD interaction interfaces and varying degrees of calcium binding dependent fluorescence change. Our results indicate that biosensor performance can be predicted based on distribution of water molecules around the chromophore and shifts in hydrogen bond occupancies between the ligand-bound and ligand-free sensor structures. Hydrogen bond occupancies were calculated with merging_bonds.py script. Double counted hydrogen bonds where a residue acts both as acceptor and donor are merged into a single entry with merge_files.py. To run sasa.tcl, you need VMD software. Trajectories were created with NAMD2 with a dcdfrequency of 5000 timesteps (every 10 ps) and strided in a 1:100 ratio (every 1 ns=1 frame in dcd).Melike Berksoz, Canan Atilgan* 萨班哲大学工程与自然科学学院 *通讯作者：卡南·阿蒂尔甘，萨班哲大学自然科学与工程学院，土耳其伊斯坦布尔图兹拉34956，电子邮箱：[e-mail] 基因编码荧光生物传感器（Genetically Encoded Fluorescent Biosensors, GEFBs）已成为活体可视化生物过程的不可或缺的工具。典型的GEFB由传感结构域（sensory domain, SD）与基因融合的荧光蛋白（fluorescent protein, FP）组成：SD在配体结合后会发生构象变化，配体与SD结合后会通过变构效应调节发色团的微环境，进而改变其光谱特性。基于单个荧光蛋白的生物传感器作为GEFB的一个子类，具备实验设置简洁的优势：易于在活细胞中表达制备、结构稳定且操作简便，仅需单波长即可完成检测。但这类传感器在结构优化方面存在显著挑战，尤其是FP与SD之间连接肽的长度和残基组成，会直接影响发色团对SD构象变化的响应效率。本研究采用经典全原子分子动力学模拟，分析了一系列基于钙调蛋白的钙生物传感器的动态特性，这些传感器具有不同的FP-SD相互作用界面，且钙结合依赖的荧光变化程度各不相同。研究结果显示，可通过发色团周围水分子的分布情况，以及配体结合与未结合状态下传感器结构的氢键占有率变化，预测生物传感器的性能。 氢键占有率通过merging_bonds.py脚本计算得到。对于同时作为氢键供体和受体的残基所产生的重复计数氢键，使用merge_files.py脚本将其合并为单一条目。运行sasa.tcl需依赖VMD软件。轨迹文件由NAMD2生成，dcdfrequency设置为5000个时间步（即每10 ps输出一次），并以1:100的比例进行步长抽取（即每1 ns对应dcd文件中的1帧）。