COVID-19 immunological profiling with flow cytometry

Source code for manuscript:

"Longitudinal immune profiling of mild and severe COVID-19 reveals innate and adaptive immune dysfunction and provides an early prediction tool for clinical progression"

Posted on medRxiv, 2020/09/09: https://doi.org/10.1101/2020.09.08.20189092

Published on Life Science Alliance, 2020/12/24: https://doi.org/10.26508/lsa.202000955 as

"Profiling of immune dysfunction in COVID-19 patients allows early prediction of disease progression"

Organization

• This CSV file contains the original manually curated data
• The metadata directory contains parsed/further curated metadata
• Raw data (e.g. CSV, FCS and H5ad files) will be under the data directory.
  • FCS files can be get from Cytobank (requires account, described below).
  • H5ad files can be get from the following URLs using the make get_h5ad command (see more below).
• The src directory contains source code used to analyze the data
• A Makefile is provided to allow easy execution of task blocks.
• Outputs from the analysis will be present in a results directory, with subfolders pertaining to each part of the analysis described below.

FCS files are hosted on Cytobank. An account is needed to download the files, which can be made programmatically with the make get_fcs command (see below). To connect with your account, simply add your credentials to a file named ~/.cytobank.auth.json containing the fields username and password:

{"username": "username", "password": "ABCD1234"}

Be sure to make the file read-only (e.g. chmod 400 ~/.cytobank.auth.json).

Reproducibility

Running

To see all available steps type:

make help

Makefile for the covid-flowcyto package/project.
Available commands:
help            Display help and quit
requirements    Install software requirements with pip
get_fcs         Download FCS files from Cytobank
get_h5ad        Download H5ad files
get_batch       Parse processing date from FCS metadata
parse           Parse original data into metadata and matrix data
impute          Imputation of missing FACS data
clinical        Run analysis of clinial data
unsupervised    Run unsupervised analysis
supervised      Run supervised analysis
temporal        Run temporal analysis
single_cell     Run single-cell analysis
all             Run all analysis steps in order

To reproduce analysis, simply do:

make requirements
make

Requirements

• Python 3.7+ (was run on 3.8.2)
• Python packages as specified in the requirements file - install with make requirements or pip install -r requirements.txt.

Virtual environment

It is recommended to compartimentalize the analysis software from the system's using virtual environments, for example.

Here's how to create one with the repository and installed requirements:

git clone [email protected]:ElementoLab/covid-flowcyto.git
cd covid-flowcyto
virtualenv .venv
source activate .venv/bin/activate
pip install -r requirements.txt