/enri

ENRI: A tool for selecting structure-based virtual screening target conformations.

Primary LanguagePython

Read the paper here

ENRI

ENRI is a tool for selecting structure-based virtual screening targets. The tool is a binary classifier coupled with a synthetic over-sampling procedure. ENRI, currently, comprises four programs: enri.py, pdb2descriptors.py, descriptors2predictions.py, plot_hist.py.

PREQUISITES

A fully funtional DogSiteSCorer. A python plotting library: matplotlib.* A python tabulation library.**

*only when plotting is desired **only when tabulation is desired

enri.py

This is the main program where all of ENRI's functionalities are defined.

pdb2descriptors.py

Extracts pockets and descriptors from pdb files. Interfaces with DoGSiteScorer. Please make sure you have fully funtional DoGSiteScorer.

INPUT: pdb_path OUTPUT: desc_merged.txt ARGUMENTS: pdb_path USAGE: python pdb2descriptors.py pdb_path EXAMPLE: python pdb2descriptors.py /enri_rc8/sample_files/pdbdir

descriptors2predictions.py

Predicts and writes an output file for top n predicted conformations. The output file is written to the input directory

INPUT: desc_merged.txt OUTPUT: predicted.txt ARGUMENTS: input_path, number of desired output (n), over-sampling paramter (beta), ranker (wp or p) USAGE: python descriptors2predictions desc_merged.txt, n, beta,ranker EXAMPLE: python descriptors2predictions.py /enri_rc8/sample_files/descdir/desc_merged.txt 10 0.5 wp

plot_hist.py

Plots histogram from a desc_merged.txt file.

INPUT: file_path OUTPUT: *descriptorname.pdf ARGUMENTS: file_path USAGE: python plot_hist.py file_path EXAMPLE: python plot_hist.py /enri_rc8/sample_files/descdir/desc_merged.txt