This repository contains the code and data used to generate results in the paper "Information Encoding and Decoding in In-Vitro Neural Networks on Micro Electrode Arrays through Stimulation Timing". The code has been cleaned and refactored to improve readability and versions of all packages have been updated since original data processing for the paper was conducted. This, and the use of methods that involve random/stochastic processes (like permutation testing) may result in slight variations in results when replicating the paper. The code is also fairly computationally heavy, to reduce computation load we advice users to reduce the n_training_rounds parameter in the notebooks that include "classification" in their title from 100 to 2-5 to see results quicker. The computer used to compute results used a 64 core CPU and 256 GB RAM. Users may need to reduce the parameter: n_workers, used when initiating the dask Client to be able to run the code successfully. For maximal speed, n_workers should equal the number of CPU cores on your machine. However, a smaller number may be needed depending on the amount of RAM available. Poetry has been used for package managment and all updated package versions can be found in the pyproject.toml and poetry.lock files under Analysis and Data/Analysis/axion_MEA_analysis. Code can be found found in the same folder. To replicate the analysis, run the jupyter notebooks found under Pipelines. Code related to each section is placed in folders with the same naming convention as the related section in the paper. Before running this code please change the path variables to where the data is placed on your own computer. The data and code in this repository is licensed under cc by-nc-nd 4.0 as described in the license file.