This folder contains python scripts and required data to perform calculations for the paper "Predicting protein-peptide interactions: benchmarking Deep Learning techniques and a comparison with focused docking ". Please cite: Shanker, Sudhanshu, and Michel F. Sanner. "Predicting protein–peptide interactions: benchmarking deep learning techniques and a comparison with focused docking." Journal of Chemical Information and Modeling 63.10 (2023): 3158-3170. DOI: https://doi.org/10.1021/acs.jcim.3c00602

The function "estimate_energies_for_pdb" from OpenMM_functions.py can be used perform minimization and various energy calculations for a protein-peptide complex.

The percentage docking success rate as a function of Fnat cutoff can be calculated by "plot_fnat_vs_success_shared.py".

To compare the performance of different methods "cross_performance_analysis.py" script can be used.

The details of PDB files, protein sequences, and peptide sequences for benchmarking is given in "pdb_list.txt" and "pdb_sequences.csv" files.

The file "all_fnat.dat" contains Fnat values from all methods for different top solutions.