Dynamic 1D search and processive nucleosome translocations by RSC and ISW2 chromatin remodelers

# Dynamic 1D Search and Processive Nucleosome Translocations by RSC and ISW2 Chromatin Remodelers'a0 Raw data of single molecule diffusion on naked DNA or on nucleosome arrays was collected on DNA stretched using dual optical tweezers and imaged for fluorescence using point scanning confocal microscopy. Initial raw data is in the form of an h5 file, which contains correlated metadata. However, due to the very large initial file sizes, we have chosen to upload images of kymographs (as tiff files) used for single particle tracking along with the respective single particle trajectories for both the red and green channel (as csv files). Frame-rates and pixel sizes are the same for all kymographs (0.0424 s and 0.1 um respectively). Single particle tracking was performed using LUMICKS Pylake Kymotracker as described in the methods section of our associated manuscript. The uploaded data can be found in three parts as zip files: (1) naked DNA data: remodeler diffusion on naked lambda DNA (2) nucleosome array data: remodeler diffusion on nucleosome arrays (3) remodeler-remodeler data: diffusion of remodelers labeled in red and green diffusing on the same naked lambda DNA The csv files uploaded contain the following data as comma delimited columns (all in column 1 of the csv file): (1)Line index (this is the trajectory number) (2)Time (pixels): to convert to seconds 1 pixel = 0.0424 s (3)Coordinate (pixels): to convert to distance 1 pixel = 0.1 um (4)Time (s) (5)Position (um) (6)Counts (summed over 7 pixels): intensity of the molecule Tiff files can be used as they are or can be converted to jpeg using software such as FIJI ## Folders/Subfolders Main Folders: "naked DNA", "nucleosome array", "remodeler-remodeler". Each main folder name describes the experimental conditions that help identify the contained data with the section of the manuscript where the data is used. For example: all diffusion data of single remodelers collected on naked DNA is in the "naked DNA folder". For single remodelers on the nucleosome array, data is in the "nucleosome array" folder. For data of two remodelers on the same piece of naked DNA, the data is in the folder "remodeler-remodeler". Subfolders: Given the description of the contents of the main folders, the user will be able to understand which data is contained in the subfolders by the description included in the name of the subfolder. For example, in the "naked DNA" main folder, the subfolder "ISW2_1mMATP" includes data of ISW2 diffusion in the presence of 1mM ATP. In the case of the "nucleosome array" main folder, the subfolder "ISW2" with subfolder "ATP" includes data on ISW2 diffusion in the presence of ATP on the nucleosome array. In the case of the "remodeler-remodeler" main folder, the subfolder "RSC-ISW2" includes data of where RSC and ISW2 are labeled in different colors and are diffusing on the same piece of DNA. ## Code/Software Source codes are written in Matlab Analysis of the resulting single particle trajectories (csv files) can be performed as follows, using the provided Matlab source code. The result of this analysis are parameter files for each single particle trajectory detected within each kymograph. These parameter files contain amongst other parameters the generalized diffusion coefficient. Further processing of these files produces summary files for each condition: '93DValuesList.csv'94: mean diffusion coefficient values per diffusive class [i.e. non-diffusive, low-diffusion, high-diffusion], '93DwellTimeList.csv'94: list of dwell-times in a diffusive state after transitioning into that diffusive class, '93Parameter_Summary.csv'94: relevant parameters used for the analysis [i.e. frame-rate, rolling-window size, threshold values differentiating diffusive classes], and '93StatePercentagesPerTrajectory.csv'94: percentage of time spent in a given diffusive class per trajectory. Run '93c01_smoothKymo.m'94 and select the folder containing the csv files (this needs to be done twice in the case of two-color channels, once for the data in the red channel and once for the data in the green channel). This step smooths the tracking data, which will be the input for subsequent analysis. 2.'a0'a0'a0'a0 Run the rolling-window analysis.'a0 This must be run once for data in the red channel and a second time for data in the green channel.'a0 Rather than selecting the folder of interest as was done in step 1, this step was designed to feed in all data folders for a given remodeler. Select the folder containing all data folders. These sub-data folders must have '93_csvfiles'94 at the end of the folder name to be detected for batch processing. a.'a0'a0'a0'a0 For data in the red color channel: run '93c02_1_noPlot_Rollingwindow_CalculateParameters_Plot_AllCSV_VER2.m'94 b.'a0'a0'a0'a0 For data in the green color channel: run '93c02_2_ALT_noPlot_Rollingwindow_Nucleosomes.m'94 3.'a0'a0'a0'a0 Finally, in the same was a was done in step 2, run: '93c03noplotCompute_DStates_fromcsv_NakedDNA_stringentsliceminimum.m'94 to create summary information for the condition of interest. Functions used in this analysis are provided in the source code folder. When running the scripts, please select change folder so that the functions can be found. Note: @msdanalyzer was used for calculating diffusion coefficients. This Matlab package needs to be downloaded and added to the same folder as the functions provided in order to run. Nadine Tarantino, Jean-Yves Tinevez, Elizabeth Faris Crowell, Bertrand Boisson, Ricardo Henriques, Musa Mhlanga, Fabrice Agou, Alain Isra'ebl, and Emmanuel Laplantine. TNF and IL-1 exhibit distinct ubiquitin requirements for inducing NEMO-IKK supramolecular structures. J Cell Biol (2014) vol. 204 (2) pp. 231-45 An example of the complete analysis is provided for the following data set: nucleosome array -> ISW2 -> ATP -> 20211117_example_csvfiles The resulting analysis is added as subfolders of the respective color channels, with each subsequent analysis creating a new subfolder.

# RSC与ISW2染色质重塑酶介导的动态一维搜索与持续性核小体易位

本数据集收录了裸DNA或核小体阵列上单分子扩散的原始数据：实验采用双光镊拉伸DNA，并通过点扫描共聚焦显微镜进行荧光成像。原始数据初始格式为含关联元数据的h5文件。由于初始文件体积过大，本数据集仅上传用于单粒子追踪的时间-空间图（kymograph）的TIFF文件，以及红绿双通道各自对应的单粒子轨迹逗号分隔值（CSV）文件。

所有时间-空间图的帧率与像素尺寸均保持一致：帧率为0.0424秒，像素尺寸为0.1微米。单粒子追踪采用LUMICKS Pylake Kymotracker完成，具体操作详见相关论文的方法部分。

上传的压缩包数据分为三部分：
1. 裸DNA数据集：染色质重塑酶在裸λDNA上的扩散数据
2. 核小体阵列数据集：染色质重塑酶在核小体阵列上的扩散数据
3. 重塑酶-重塑酶数据集：分别被红绿荧光标记的两种重塑酶在同一条裸λDNA上的扩散数据

上传的CSV文件包含以下逗号分隔的列数据：
1. 行索引：即轨迹编号
2. 时间（像素单位）：转换为秒的换算系数为1像素=0.0424秒
3. 坐标（像素单位）：转换为距离的换算系数为1像素=0.1微米
4. 时间（秒）
5. 位置（微米）
6. 计数（7像素累计值）：对应分子的荧光强度

TIFF文件可直接使用，或通过FIJI等软件转换为JPEG格式。

## 文件夹与子文件夹
主文件夹共3个，分别为「裸DNA」、「核小体阵列」、「重塑酶-重塑酶」。每个主文件夹的名称对应对应的实验条件，可用于匹配数据在论文中的对应章节。例如：所有在裸DNA上采集的单染色质重塑酶扩散数据均存放于「裸DNA」文件夹；在核小体阵列上采集的单染色质重塑酶扩散数据存放于「核小体阵列」文件夹；两条分别标记的重塑酶在同一条裸DNA上的扩散数据则存放于「重塑酶-重塑酶」文件夹。

子文件夹：结合主文件夹的内容说明，用户可通过子文件夹的命名快速识别其内部包含的数据。例如：在「裸DNA」主文件夹中，子文件夹「ISW2_1mMATP」存放了1mM ATP存在条件下ISW2的扩散数据；在「核小体阵列」主文件夹中，子文件夹「ISW2」下的「ATP」子文件夹存放了ATP存在条件下ISW2在核小体阵列上的扩散数据；在「重塑酶-重塑酶」主文件夹中，子文件夹「RSC-ISW2」存放了分别被红绿荧光标记的RSC与ISW2在同一条DNA上的扩散数据。

## 代码与软件
本数据集附带的分析代码均采用Matlab编写。可通过以下步骤，使用附带的Matlab源代码对CSV格式的单粒子轨迹数据进行分析。

对轨迹数据的分析结果为每个时间-空间图中检测到的单粒子轨迹对应的参数文件，除其他参数外，此类文件包含广义扩散系数。对参数文件进行进一步处理后，可生成对应实验条件的汇总文件：
1. `DValuesList.csv`：各扩散类别（即非扩散、低扩散、高扩散）的平均扩散系数
2. `DwellTimeList.csv`：各扩散类别的驻留时间列表（即进入该扩散状态后在该状态下停留的时长）
3. `Parameter_Summary.csv`：分析所用的相关参数（包括帧率、滑动窗口大小、区分扩散类别的阈值等）
4. `StatePercentagesPerTrajectory.csv`：每条轨迹在各扩散类别中花费的时间占比

1. 运行`c01_smoothKymo.m`，选择包含CSV文件的文件夹（双通道数据需执行两次该步骤：分别处理红色通道与绿色通道的数据）。该步骤将对追踪数据进行平滑处理，作为后续分析的输入。
2. 执行滑动窗口分析。该步骤同样需分别处理红色与绿色通道的数据，且无需像步骤1那样单独选择目标文件夹，而是选择包含所有待分析数据文件夹的根目录。需注意：待处理的子数据文件夹名称必须以`_csvfiles`结尾，方可被批量处理流程识别。
a. 针对红色通道数据：运行`c02_1_noPlot_Rollingwindow_CalculateParameters_Plot_AllCSV_VER2.m`
b. 针对绿色通道数据：运行`c02_2_ALT_noPlot_Rollingwindow_Nucleosomes.m`
3. 最后，参照步骤2的操作方式，运行`c03noplotCompute_DStates_fromcsv_NakedDNA_stringentsliceminimum.m`，生成对应实验条件的汇总信息。

分析过程中用到的函数均存放于源代码文件夹中，运行脚本前请先切换工作目录至该文件夹，确保函数可被正确调用。
注意：扩散系数的计算采用了`msdanalyzer`工具包，需提前下载该Matlab工具包，并将其添加至与附带函数相同的工作目录中，方可正常运行分析代码。

### 参考文献
Tarantino N, Tinevez J-Y, Crowell EF, Boisson B, Henriques R, Mhlanga M, Agou F, Israël A, Laplantine E. TNF与IL-1诱导NEMO-IKK超分子结构的泛素依赖机制存在差异. J Cell Biol. 2014;204(2):231-245.

本数据集提供了完整分析流程的示例，对应数据集为「核小体阵列」→「ISW2」→「ATP」→「20211117_example_csvfiles」。分析结果将作为对应颜色通道的子文件夹存放，每一轮后续分析都会生成新的子文件夹。