This repository contains our code to find candidate markers for cell-type specific exosome isolation from the brain.
Broadly, the code uses protein annotations, gene expression data, and proteomics data. We isolate candidate proteins or genes which have transmembrane domains, are specifically expressed in a desired cell-type in the brain, and are specifically expressed in the brain compared to other tissues. Then, we compare to our own mass-spectrometry data of CaptoCore-isolated exosomes from plasma, CSF, and iPS neuron culture, as well as past datasets from the human plasma proteome project (HPPP) and others.
The data can be found at https://drive.google.com/file/d/1UszDdikP1j9NT8kU1qZmKsViOnuZB3BN/view?usp=sharing. Download this to the same folder where you clone the repository and unzip it there. Make a new folder called 'results' as well.
The structure of the repository when set up should be as follows:
neuron_EV_markers/
├── bin/ # Scripts and executables
├── data/ # Input data files
├── figures/ # Figures for the manuscript
├── results/ # Analysis outputs
├── supp_tables/ # Supplementary tables
./bin/ contains the code for the analysis, ./data/ contains the raw data used in the analysis, ./figures/ contains the figures produced for the paper, ./results contains other figures and results of the code which we found useful to save, and ./supp_tables/ contains the supplementary tables produced for the paper.
The major Python package requirements and their tested versions are in requirements.txt.
This code was run with Python version 3.7.
To replicate our results, please clone this repository, download the data from the Google Drive link and install the python packages denoted in the requirements.txt file. Then, use the ./bin/figure_producer.ipynb notebook and click "Run All".
If you would like to play around with our code or generate your own lists of candidates we have made a notebook called ./bin/sandbox.ipynb.
This should output figures in the ./results folder. The figures which were specifically used in the paper should be in the ./figures folder, and supplemental tables should be in the ./supp_tables folder.
Please let us know if you have difficulty using this!
Siddharth Iyer
For questions, please use the GitHub Discussions forum. For bugs or other problems, please file an issue.