README - FILE
Grasp - GRaSP: a graph-based residue neighborhood strategy to predict binding sites
------------------------------------
Charles Abreu Santana,
Universidade Federal de Minas Gerais,
Belo Horizonte, Minas Gerais, Brasil.
VERSION 1.0.0 May 2020.
GRaSP is a residue centric method to predict ligand biding site residues. It is
based on a supervised learning strategy that models the residue environment as a
graph at the atomic level. The program is written in python and should work on
most UNIX like environment.
usage
-----
The program is run by the python script called "grasp.py". To use GRaSP, simply
type "python3 grasp.py", using the parameters -p followed by the name of a valid
PDB file or a directory contaning many PDB valid files, and -o followed by the
name of a output directory to save the results, as shown below
python3 grasp.py -p proteinFile -o outputDir
or
python3 grasp.py -p proteinDir -o outputDir
By default, GRaSP uses the exposure information calculated by Biopython library
to perform the preditcion. You could use the naccess program to achieve more
accurated results (http://wolf.bms.umist.ac.uk/naccess/). If you want to use it
just type -n followed by the directory name where naccess is, as shown below.
python3 grasp.py -p proteinFile -o outputDir -n naccessDir
example output files
--------------------
The output is a .csv file containing two columns: the first one (res_name) which
contains the residue information and the second one (prediction) with the class
label predicted. The class label is binary. Class equal to 1 represents positive
(residue biding site) and 0 means negative (not residue biding site). The residue
infomation is composed by pdb Id, chain, residue name and residue number joined by
underlines.
res_name,prediction
1gkc_A_TYR_420,0
1gkc_A_PRO_421,1
1gkc_A_MET_422,1
1gkc_A_TYR_423,1
1gkc_A_ARG_424,0
1gkc_A_PHE_425,0
1gkc_A_THR_426,0
1gkc_A_GLU_427,0
1gkc_A_GLY_428,0
1gkc_A_PRO_429,0
1gkc_A_PRO_430,0