https://github.com/bionetslab/robust
ROBUST: Robust disease module mining via enumeration of price-collecting Steiner trees
https://github.com/bionetslab/robust
algorithm
Last synced: 23 days ago
JSON representation
ROBUST: Robust disease module mining via enumeration of price-collecting Steiner trees
- Host: GitHub
- URL: https://github.com/bionetslab/robust
- Owner: bionetslab
- License: gpl-3.0
- Created: 2021-06-28T13:10:55.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2024-02-06T15:55:53.000Z (almost 2 years ago)
- Last Synced: 2024-02-06T16:54:04.782Z (almost 2 years ago)
- Topics: algorithm
- Language: Jupyter Notebook
- Homepage:
- Size: 8.71 MB
- Stars: 5
- Watchers: 3
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# New ROBUST version
We have developed a new version of ROBUST which corrects for study bias in PPI networks. The new version is available here: https://github.com/bionetslab/robust_bias_aware. Although this version is still fully functional, we recommend using the latest ROBUST version for your research.
# Installation
Install conda environment as follows (there also exists a environment.yml but it contains more packages than necessary)
```bash
conda create --name biosteiner python=3.7
conda activate biosteiner
conda install numpy matplotlib pandas networkx pip jupyter
pip install pcst_fast
```
# Running ROBUST
You can simply run robust by calling
```bash
python robust.py data/human_annotated_PPIs_brain.txt data/ms_seeds.txt ms.graphml 0.25 0.9 30 0.1
```
The positional arguments are:
```
[1] file providing the network in the form of an edgelist
(tab-separated table, columns 1 & 2 will be used)
[2] file with the seed genes (if table contains more than
one column they must be tab-separated; the first column
will be used only)
[3] path to output file
[4] initial fraction (alpha)
[5] reduction factor (beta)
[6] number of steiner trees to be computed
[7] threshold (theta)
```
The suffix of the path to the output file you specify, determine the format of the output.
You can either choose
- .graphml: A .graphml file is written that contains the following vertex properties: isSeed, significance, nrOfOccurrences, connected_components_id, trees
- .csv: A .csv file which contains a vertex table with #occurrences, %occurrences, terminal (isSeed)
- everything else: An edge list
# Evaluating ROBUST
For a large-scale empirical evaluation of ROBUST, please follow the instructions given here: https://github.com/bionetslab/robust-eval.
# Citing ROBUST
Please cite ROBUST as follows:
- J. Bernett, D. Krupke, S. Sadegh1, J. Baumbach, S. P. Fekete, T. Kacprowski, M. List1, D. B. Blumenthal: Robust disease module mining via enumeration of diverse prize-collecting Steiner trees, *Bioinformatics* 38(6), pp. 1600-1606, https://doi.org/10.1093/bioinformatics/btab876.