Benchmark_Instructions

### Benchmark
Promotech was compared against previous developed models. The models' linux executable files are located at the `./models/preceding/` folder. The instructions to run each program are the following:

1. **BPROM** 

Source: http://linux1.softberry.com/berry.phtml?topic=fdp.htm

- Setup and run the Ubuntu 16.04 32bits docker container
   - `docker pull 32bit/ubuntu:16.04`
   - `cd models/preceding/bprom`
   - `docker run -it --name ubuntu32 -v $(pwd):/project 32bit/ubuntu:16.04 /bin/bash`
- Run the program
   - `export TSS_DATA="/project/bprom_data/"`
   - `/project/linux/bprom /project/example_data/seq.fa /project/out.txt`
- Check the results
   - `head /project/out.txt`
- Stop the docker container
   - `docker rm ubuntu32`

         
2. **bTSSfinder**

Source: https://www.cbrc.kaust.edu.sa/btssfinder/about.php
Article: https://doi.org/10.1093/bioinformatics/btw629

- Setup and run the Ubuntu 18.04 64bits docker container
   - `docker pull ubuntu/18.04`
   - `cd models/preceding/btssfinder`
   - `docker run -it --name ubuntu64 -v $(pwd):/project ubuntu:18.04 /bin/bash`
- Run the program
   - `sudo chmod +x /project/bTSSfinder`
   - `export bTSSfinder_Data="/project/Data/"`
   - `project/bTSSfinder -i /project/example.fasta -o /project/out`
- Check the results
   - `head /project/out.txt`
- Check the results. The program will generate two files called `out.bed`, `out.gff`, and `out.out`.

3. **G4Promfinder**

Source: https://github.com/MarcoDiSalvo90/G4PromFinder
Article: https://doi.org/10.1186/s12859-018-2049-x

- Setup libraries

   - `pip3 install numpy pandas matplotlib biopython openpyxl`

- Run the program
    - `cd models/preceding/g4promfinder`
    - `python3 g4promfinder.py`
- The program will ask to input the name and location of the FASTA file. Inside the `models/preceding/g4promfinder` folder, there is an example file called `example.fasta` you can use to test the program.
   
   ```G4PromFinder is a genome-wide predictor of transcription promoters in bacterial genomes. It analyzes both the strands. It is recommended for GC-rich genomes
   Input: Genome file
   Output: a text file containing promoter coordinates and a file containing informations about them
   Genome file must be in fasta format
   Enter the input genome file name: example.fasta
   G4PromFinder is working, please wait...
   Work finished, see output files in the current directory
   ```

- Check the results. The program will generate two files called `ABOUT PROMOTERS.txt` and `promoter coordinates.txt`.

4. **MULTiPLy**

Article: https://doi.org/10.1093/bioinformatics/btz016

- Download and install the latest version of Matlab. https://uk.mathworks.com/downloads/web_downloads/
- `cd models/preceding/multiply`
- Edit the `MULTiPLy.md` file and change the 12th line, `[head,seq]=fastaread('sample.fasta');`, with the fasta file you want to test. You can use `sample.fasta` for testing.
- Run the `MULTiPLy.md` file from the Matlab's GUI command line.

         
5. **iPromoter-2L**

Article: https://doi.org/10.1093/bioinformatics/btx579
Data: http://bioinformatics.hitsz.edu.cn/iPromoter-2L/data/

`python promotech.py -b -bm iPromoter2L -m RF-HOT -o results`