https://github.com/tcp-lab/transportome_profiler
Code to profile the expression of the transportome
https://github.com/tcp-lab/transportome_profiler
Last synced: 7 months ago
JSON representation
Code to profile the expression of the transportome
- Host: GitHub
- URL: https://github.com/tcp-lab/transportome_profiler
- Owner: TCP-Lab
- License: gpl-3.0
- Created: 2023-03-15T10:52:25.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2025-02-17T15:53:57.000Z (over 1 year ago)
- Last Synced: 2025-02-17T16:41:31.812Z (over 1 year ago)
- Language: R
- Homepage:
- Size: 2.13 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Transportome Profiler analysis
This repository contains the code for the analysis on the expression profile of
the transportome in Cancer based on the [MTP-DB](https://github.com/TCP-Lab/MTP-DB).
> [!NOTE]
> Read the preprint here: [Profiling the Expression of Transportome Genes in cancer: A systematic approach](https://doi.org/10.1101/2023.07.18.549498)
>
> It's potentially out of date.
The project follows the [Kerblam! standard](https://github.com/MrHedmad/kerblam).
## Running the analysis
You can run the analysis pipelines with Kerblam! and `docker`:
```bash
# Clone the repo
git clone git@github.com:TCP-Lab/transportome_profiler.git
cd ./transportome_profiler
kerblam data fetch # Fetch the input data not present in the repository
kerblam run
```
Kerblam! will build docker containers and run the analysis locally.
To run without docker, read below.
### Pipelines
The project currently encompasses the following pipelines:
- `heatmaps`: Create large heatmaps from the expression matrices by using GSEA
on computed gene rankings, testing all possible gene lists that can be made
from the `MTP-DB`.
- The `test` profile makes this pipeline much faster by running on smaller
(i.e. sampled) input data (~75% reduction in sample number, only 5000 random genes).
### Running locally without docker
You need some requirements to be installed before you can run the analysis locally:
- `R` version `4.3.0+`.
- Install R requirements with `./src/helper_scripts/install_r_pkgs.R`.
- `Python` version `3.11+`.
- Install python requirements with `pip install -r ./src/requirements.txt`.
- The `jq` utility (that you can find [here](https://jqlang.github.io/jq/)).
- The `xsv` program, required by [`metasplit`](https://github.com/MrHedmad/metasplit)
(`sudo pacman -Syu xsv` on Arch, not packaged by Debian, but [this guide might be useful](https://lindevs.com/install-xsv-on-ubuntu).
If you have `cargo` installed, you can simply run `cargo install xsv`).
- Follow the extra installation guide for [`generanker`](https://github.com/TCP-Lab/gene_ranker) (namely installing [fast-cohen](https://github.com/MrHedmad/fast-cohen))
- The `xls2csv` utility (on arch `yay -Syu perl-xls2csv`)
- A series of R packages that can be installed with `Rscript ./src/helper_scripts/install_R_pkgs.R`
- Quite a bit of RAM (some steps require > 50 Gb of RAM) and time.
Override `N_THREADS` (with `export N_THREADS=...`) to run with less threads.
If you have all the requirements, you can simply:
```bash
kerblam run --local
```
> [!IMPORTANT]
> The manuscript for this project is also available online in [this repository](https://github.com/TCP-Lab/transportome_profiler_paper).