Bio-informatics common repository for antimicrobial peptides design
The following code block allows you to setup a virtual environment on your Windows laptop. If you don't want to use a virtual environment, you can install your packages globally with the last line of the code block.
python -m venv .venv
source .venv/Scripts/activate # or .\.venv\Scripts\activate.bat if you are using Windows cmd
pip install -r requirements.txt
Link to the original repo: https://github.com/alxdrcirilo/RamachanDraw
Positive_db obtained from DRAMP server and only alpha-helixes peptides were selected : http://dramp.cpu-bioinfor.org/ Positive_db_reac_IC50 from 2005 DOI:
Peptide of short size were downloaded on Uniprot server : https://www.uniprot.org/ They were previously filtered for intracellular only presence as it was assumed they would not biologically have a relevant effet againt bacteria and fungy.
Performs database curation (positive and negative) using a size restriction filter : only AMPs in size range 3 to 18 aa are selected. It creates 2 temporary directory (pos and neg) each containing a kmr file for each peptide sequence containing all possible kmer of size 5 with maximum 3 gaps All generated kmers are then concatenated and counted to a dictionnary prior scoring.
Scoring function based on occurence of descriptor in positive relative to negative database is computed as bellow :
The score is added to each key: value couple of the dictionnary and the global data are saved in file unique_set.tsv
Analyse frequency of aa in database for given positions
Computes physico-chemicals properties of peptides based on their sequences
- net charge at pH
- plot hydrophobicity and autocorrelation transformation
- spaces between hydrophilic domains
- helical distribution of aa along helixes
Uses unique_set.tsv file to attribute a score to a given peptide sequence according to descriptors scores computed with kmer_parser.py