########################################################################
	A set of data and scripts for analysis of RAMD dissociation trajectories generated for 
   	 1. mAChR M2 muscarinic receptor bound to 
        	(i) iperoxo -IXO-CHL.zip
        	(ii) iperoxo and PAM - IXO-ALO-CHL.zip
       		(iii) ACh - ACh-CHL.zip
    	2. β2AR - β2-adrenergic receptor bound to 
        	alprenolol - b2AR.zip

##########################################################################
	Scripts were used for analysis reported in the manuscript
		"G-Protein Coupled Receptor-Ligand Dissociation Rates and Mechanisms from tauRAMD Simulations" 
 		by Daria B. Kokh, Rebecca C. Wade
 		submitte to JCTC , June 2021

##########################################################################    
	Author: Daria Kokh
    	daria.kokh@h-its.org or mcmsoft@h-its.org

    	02.06.2021
    	Copyright (c) 2020
    	Released under the EUPL Licence, v1.2 or any higher version

    	Heidelberg Institute of Theoretical Studies (HITS, www.h-its.org)
    	Schloss-Wolfsbrunnenweg 35
    	69118 Heidelberg, Germany

##########################################################################
	Packages required (version used for testing):
		numpy (v 1.18.1)
		matplotlib (3.1.3)
		sklearn (0.22.1)
		scipy (1.4.1)
		pandas (1.0.2)
		seaborn (0.10.0)
		MDAnalysis (0.20.1)
		rdkit (2019.09.3)
		code is written on Python 3.x and tested on the version 3.7

		Functions collected in the directory Scripts (Scripts.zip file)

##########################################################################

	Protocol:
	
	1. RAMD dissociation trajectories were generated using Gromacs GROMACS/2020.3-RAMD-1.0
		trajectories are not included in the present data set, but can be provided upon request

	2. P-L IFP data are generated for the last 700 frames of each RAMD trajectory and stored in *pkl DataFrame files. For this
		(i) First, each trajectory is pre-processed using Gromacs tools to wrap system into box. This can be done in automated way using the script IFP_preprocess_Gromacs.py
			as: "python ./IFP_preprocess_Gromacs.py TREX  $n > out"
			( n - replica number, TREX* - standard names of a directory containing a particular RAMD trajectory)
			this script will generated from each trajectory a new short (< 1000 frames) one
		(ii) Then P-L IFPs are generated using IFP_SL-B2AR-WB-EX.py 
			as: "python  IFP_SL-B2AR-WB-EX.py  $n  > b2AR-unwtapped-WB-EX-700-RAMD__$n.dat"
			this sctipt will generate pkl and dat files - both are needed for further analysis 
                Note, that scripts are organized for the specific directory/file structure and names for β2-adrenergic receptor.
		For any other system, thay have to be adapted accordingly
		I had to split all trajectories for each system into groups to fit in 24h of simulations.
		While *pkl file contains all TFP information, *dat containes all output informatio. 
		In particular, full path for each trajectory priocesses. This information can be useful if one has to find a particular snapshot for visualization of protein/ligand position.
	  An example is given in b2AR-alprenolol directory, where  IFP_b2AR-EX.sh can be used to run the first and the second  steps as described above

	3. Residence times are computed using the script tauRAMD-v2.py
	4. P-L IFP data are analized using the Gromacs-IFP-GPCR.ipynb Jupyter Notebook
		JN script uses also library of scripts collected in Scripts.zip
		Fot this one need to store data in a directory DATA/GPCR-09-2020 (if they are in another directory, one has to change directory name in the JN accordingly)
		Specifically,
			IXO-ALO-CHL.zip contanes *dat and *pkl files generated for mAChR M2 - iperoxo + PAM
			IXO-CHL.zip - mAChR M2 - iperoxo 
			ACh-CHL.zip - mAChR M2 - ACh
			b2AR.zip  alprenolol - b2AR
		Each archive containes a *pkl and *dat files. *pkl contane IFPs for a set of trajectories. In JN they are combined all togeather for analysis
		Additionally, there is an archive water.zip that contains water-1000_EX.dat file - number of water and ligand atoms in the binding pocket for all frames for alprenolol - b2AR system

	5. Additional plots (including plot of the residence times computed vs experimental) were generated using Auxi_Plots-GPCR.ipynb Jupyter Notebook