Skip to content

stjudecloud/expression-classification

Repository files navigation

Interactive t-SNE

Code used to generate interactive t-SNE plots using RNA-Seq counts data.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.

Prerequisites

To get started, you'll want to have Python 3 and R 3.6.1.

conda create -n itsne-dev \
             -c conda-forge \
             -c bioconda \
             r-base=3.6.1 \
             python=3.7 \
             logzero \
             numpy \
             pandas \
             scikit-learn \
             bioconductor-biocinstaller \
             bioconductor-sva \
             bioconductor-deseq2 \
             r-tsne r-getopt r-plotly r-optparse r-rtsne r-pracma r-data.table -y
conda activate itsne-dev

Alternatively, you can install the anaconda dependencies directly from the environment.yml file.

conda env create -f environment.yml

Installing

First, the author recommends you run the following command line script to ensure all R packages are loaded before your first run:

Rscript scripts/itsne-normalize-matrix.R

From here, you can install the python package:

python3 setup.py install

Download the relevant gene model:

curl ftp://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_31/gencode.v31.annotation.gtf.gz -o gencode.v31.annotation.gtf.gz

And try out the script:

itsne-main counts/* \
           -b reference/gene.excludelist.tsv \
           -c reference/covariates.tsv \
           -g gencode.v31.annotation.gtf.gz \
           -o tsne.html

Running the tests

No tests currently exist. If and when tests are added, the authors will fill in this section.

Contributing

Please read CONTRIBUTING.md for details on our code of conduct, and the process for submitting pull requests to us.

Versioning

We use SemVer for versioning. For the versions available, see the tags on this repository.

License

This project is licensed under the MIT License - see the LICENSE.md file for details