Conjunto de datos de: Packet information encoding in a cerebellum-like circuit.

DATA FILES These complementary files contain 3 sets of data (*.mat files) and 6 processing codes (*.m files) supporting our results. The data are binary files in “matlab/octave format”. The codes can be opened with any text reader (for example notepad of Windows). Two of these codes are ad hoc functions: JSdistance.m and comparison.m. The other 3 are the codes for data processing: analysis_without_object.m, makefig5.m, and step_analysis.m. PROCEDURE FOR REPRODUCING THE RESULTS OUT OF THE POINT PROCESSES Please after download, make a separate directory in your computer, move all data and code files to it.Then you can e run all processes typing 'automaticrun' in the prompt. The computer will do everything without any other required maneouver. Otherwise make two additional directories inside the main directory; naming them transient and figures.Then you can run the different codes separately as ‘m functions’ just typing in the console the names of the three main files mentioned above. This will let you to reproduce the results from the data files. It will also save different partially processed files in the folder 'transient', and save the draft figures (in Matlab/Octave) format in the folder 'figures' and it will also display tables and comments in the console. Content of the data files is explained next. BASIC PROCESSING Contains the file 'data_without_object.mat' that will be used by the function 'analysis_without_object.m' and will perform the cross-covariance, the hierarchical cluster analysis, and the analysis of unit location within the electrosensory lobe. “Data_without_object.mat” contains two cells corresponding to the spike timestamps and the EOD timestamps and a variable containing the unit locations. MOVEMENT PROCESSING The file 'moving.mat' will be used by the function 'makefig5.m' and will show the figures for analysis of movement. “Moving.mat” contains 11 cells and one variable with the titles of the figures. The cells Xforward, Yforward, Xbackward, Ybackward, contain the smoothed position of the plotter for each EOD when the object was moved from rostral to caudal or from caudal to rostral respectively. The cells distanceforward, and distancebackward contains the calculated distance traveled along the skin in either direction calculated using the Pithagoric rule (i.e. the distance traveled between two time stamps equal to the square root of the sum of the squares of the distances traveled along each axes of the plotter). The variable 'EODbaseline' is the timing of the EOD in the absence of objects. The cell Spike_Latency_backward and Spike_Latency_forward show the spike time stamps after every EOD when the object was moved in either direction and the cells Spike_Latency_baseline and EOD baseline correspond to these latencies when the fish is resting without object. STEP PROCESSING The file 'stepdata.mat' will be used by the function step_analysis.m and will show the figures in the folder figures and the tables for analysis of the effects of steps in amplitude on spike rate and spike postEOD distribution. Contains two cells and two variables. The cells UP and DO contain the files corresponding to 32 units explored with increasing and decreasing steps respectively. The cell for each unit is a matrix with one dimension of 51 corresponding to the EOD order from -25 EODs before the step increment to the 25 EODS after the step increment. The other dimension contains the spike time stamps corresponding to each ordinal EOD in any trial (i.e. time stamps corresponding to every ordinal EOD pooled from all trials). N corresponds to the number of trials and Type correspond to the unit type in the same order as in the other variables. In Type, the code 1 corresponds to sharp monomodal units, 2 corresponds to deeply inhibited units, 3 corresponds to broad monomodal units, 4 corresponds to mildly inhibited units, 5 correspond to bimodal units and 6 corresponds to trimodal units.

数据文件
本配套文件包含3组*.mat格式数据文件与6份*.m格式处理代码，可用于复现本文的研究结果。所有数据均为Matlab/Octave格式的二进制文件；处理代码可通过任意文本阅读器（如Windows记事本）打开。其中2份为专用函数：JSdistance.m与comparison.m，剩余3份为数据处理代码：analysis_without_object.m、makefig5.m及step_analysis.m。

点过程结果复现流程
下载完成后，请在本地计算机新建独立目录，将所有数据与代码文件移入该目录。随后在命令行输入automaticrun即可运行全部流程，无需其他手动操作。若需分步运行各流程，可在主目录内新建两个子目录，分别命名为transient与figures，随后在控制台输入前述3份主代码文件的名称，以M函数形式分别运行各模块，即可基于数据文件复现全部研究结果。运行过程中，程序会将部分中间处理文件保存至transient文件夹，将Matlab/Octave格式的草稿绘图文件保存至figures文件夹，并在控制台输出统计表格与相关注释。

数据文件内容说明
下文将对各数据文件的内容进行详述。

基础处理模块
该模块对应文件为data_without_object.mat，将由analysis_without_object.m函数调用，用于完成电感觉叶内的互协方差分析、层次聚类分析以及神经元单元位置分析。data_without_object.mat包含两个元胞数组，分别对应脉冲发放时间戳与电官放电（Electric Organ Discharge, EOD）时间戳，另有一个存储神经元单元空间位置的变量。

运动处理模块
文件moving.mat将由makefig5.m函数调用，用于生成运动分析相关可视化图表。moving.mat包含11个元胞数组与一个存储图表标题的变量。元胞数组Xforward、Yforward、Xbackward、Ybackward分别对应物体从吻侧向尾侧、尾侧向吻侧移动时，绘图仪在每个EOD周期下的平滑位置数据。元胞数组distanceforward与distancebackward存储了基于勾股定理计算得到的皮肤表面沿各方向的移动距离（即两个时间戳间的移动距离为绘图仪各轴移动距离平方和的平方根）。变量EODbaseline为无物体存在时的EOD时间序列。元胞数组Spike_Latency_backward与Spike_Latency_forward分别存储物体沿两个方向移动时，每个EOD周期后的脉冲发放时间戳；元胞数组Spike_Latency_baseline与EOD_baseline对应无物体时鱼体静止状态下的上述潜伏期数据。

阶跃处理模块
文件stepdata.mat将由step_analysis.m函数调用，用于生成figures文件夹内的可视化图表，并输出阶跃幅度对脉冲发放率及EOD后脉冲分布影响分析的统计表格。该文件包含两个元胞数组与两个变量。元胞数组UP与DO分别对应32个经递增阶跃与递减阶跃测试的神经元单元的数据。每个单元对应的元胞为一个矩阵：其第一维度长度为51，对应阶跃变化前后的EOD序号（从阶跃变化前25个EOD至阶跃变化后25个EOD）；第二维度存储每个有序EOD对应的脉冲发放时间戳（即从所有试验中汇总得到的每个有序EOD对应的时间戳）。变量N为试验总次数，变量Type则按与其他变量一致的顺序存储神经元单元的类型信息。其中，代码1对应尖锐单峰单元，2对应深度抑制型单元，3对应宽调谐单峰单元，4对应轻度抑制型单元，5对应双峰单元，6对应三峰单元。