Cofactor-dependent feature generation for protiens
git clone https://github.com/ESPhoenix/coFeatures
conda create -n coFeaturesEnv python=3.7.12
conda activate coFeaturesEnv
pip install argpass==0.0.2 numpy==1.21.6 pandas==1.3.5 tqdm==4.66.1
import numpy as np
def inputs():
## LOCATION OF YOUR PDB FILES
inputDir = "/home/esp/featureGenerationMethods/coFeatures/test_pdbs"
## WHERE DO YOU WANT YOUR OUTPUTS
outDir = "/home/esp/featureGenerationMethods/coFeatures/features"
## LOCATION OF AA PROPERTIES
aminoAcidTable="/home/esp/featureGenerationMethods/coFeatures/amino_acid_properties.txt"
# cofactor information
## LIST OF COFACTOR NAMES IN YOUR PDB FILES
cofactorNames = ["FAD", "FMN"]
## DICT CONTAINING: RESIDUE_NAME:[ATOM1,ATOM2...]
## WILL BE USED TO GENERATE NEAREST 1 and NEAREST 3 RESIDUES FEATURES
keyAtomsDict = {
"FMN":["N1","N3","N5"],
"FAD":["N1","N3","N5"]
}
## DICT CONTAINING: RESIDUE_NAME:[ATOM1,ATOM2...]
## WILL BE USED TO DEFINE CENTER OF ORB REGION
orbAtomsDict = {
"FMN":["N10","C10","C4A","N5","C5A","C9A"],
"FAD":["N10","C10","C4X","N5","C5X","C9A"]
}
## DICT CONTAINING: RESIDUE_NAME:[ATOM1,ATOM2...]
## WILL BE USED TO DEFINE ATOMS TO CONTRUCT CLOUD REGION
cloudAtomsDict = {
"FMN":["N10","C9A","C9","C8","C8M","C7","C7M","C6","C5A","N5",
"C4A","C4","O4","N3","HN3","C2","O2","N1","C10"],
"FAD":["N10","C9A","C9","C8","C8M","C7","C7M","C6","C5X","N5",
"C4X","C4","O4","N3","HN3","C2","O2","N1","C10"]
}
## LIST OF RADII FOR ORB REGION
orbRange = list(np.arange(6, 10))
## LIST OF RADII FOR CLOUD REGION
cloudRange = list(np.arange(3, 7))
return (inputDir, outDir, aminoAcidTable, cofactorNames, keyAtomsDict,
orbAtomsDict, cloudAtomsDict, orbRange, cloudRange)
From the coFeatures directory, run the following in the command line:
python coFeatures.py --config config.py
You may want to change the name of config.py to your own config file