#########################################
############### PCA ###################
#########################################
import matplotlib as mpl
mpl.use('Agg')
import os
import numpy as np
from sklearn.decomposition import PCA
import pickle, sys
from sklearn.manifold import TSNE
# from plot_score_correct import *
from extract_pts import *
import matplotlib.pyplot as plt
from rdkit.Chem import MolFromSmiles
from rdkit import Chem, DataStructs
from rdkit.Chem import AllChem
from io_utils_DUDE_neg_poc import get_all_mols
from io_utils_MUV_ROC import *
import numpy
import os
import matplotlib as mpl
mpl.use('Agg')
import matplotlib.pyplot as plt

# from mpl_toolkits.mplot3d import Axes3D

def save(path, ext='png', close=True, verbose=True):

	directory = os.path.split(path)[0]
	filename = "%s.%s" % (os.path.split(path)[1], ext)
	if directory == '':
		directory = '.'

	if not os.path.exists(directory):
		os.makedirs(directory)
 
	savepath = os.path.join(directory, filename)
 
	if verbose:
		print("Saving figure to '%s'..." % savepath),

	plt.savefig(savepath, dpi=900)
	if close:
		plt.close()
 
	if verbose:
		print("Done")

def get_FF(target,PDB_ID,DUDE_ctr):
	print (target, PDB_ID, DUDE_ctr)
	pro_name=target
	ff_name = None
	ligs=cut_ligand_all_atoms(PDB_ID,DUDE_ctr,False)
	if ligs == []:
		print('cant find ligs'+'\n')
	for l in ligs:
		[lig_ID,lig_chain,lig_no, ctr]=l  
		#ff_name=PDB_ID+'_'+str(lig_ID)+'_'+str(lig_chain)+'_'+str(lig_no)+'.ff'
		ff_name=PDB_ID+'_'+str(lig_ID)+'.ff'
	return ff_name

def get_target_ff_dict():

	filename = "../data/DUDE/DUDE_PDBID.csv"
	data = read_csv_DUDE(filename,"Target Name","PDB")
	target_FF_dict={}
	for i in range(0,len(data[0])):
		# For each target
		target = data[0][i]
		PDB_ID = data[1][i]
		crystal_lig = open('../data/DUDE/'+target.lower()+'/'+'crystal_ligand.mol2')
		lig_ctr=parse_crystal_lig(list(crystal_lig))
		target_FF = get_FF(target,PDB_ID,lig_ctr) # Target FF
		if target_FF!=None:
			target_FF_dict[target_FF]=target
	return target_FF_dict

def get_silhouette():

	import json
	import sklearn
	from sklearn import metrics
	silhouette_dict={}

	
	with open('../target_FF_dict','r') as infile:
		target_FF_dict = json.load(infile)
	
	MUV_target = '689'
	target_lig=load_MUV_dataset()
	all_targets={}
	
	for t in target_lig:
		all_actives = []
		all_decoys = []
		
		per_targets=[]
		FOUND=False
		pro_name=t[0]
		#if pro_name=='689':
		# if pro_name==MUV_target:
			# active_file = open('../poc2vec_sda/MUV_mol/mol_order_morgan_'+pro_name+'_'+DUDE_tar+'.txt','w')
		
		PDB_ID=t[1]
		ff_name=t[2]
		pos_ligs=t[3]
		neg_ligs=t[4]
		num_neg = len(neg_ligs)
		num_pos = len(pos_ligs)
		for m in pos_ligs:
			all_actives.append(m)
		for m in neg_ligs:#[0:200]:
			all_decoys.append(m)


		
		act_mols = [ MolFromSmiles(m) for m in all_actives]
		act_fps = [AllChem.GetMorganFingerprintAsBitVect(m, 2) for m in act_mols]

		dec_mols = [ MolFromSmiles(m) for m in all_decoys]
		dec_fps = [AllChem.GetMorganFingerprintAsBitVect(m, 2) for m in dec_mols]


		# convert the RDKit explicit vectors into numpy arrays
		np_act_fps = []
		for fp in act_fps:
		  arr = numpy.zeros((1,))
		  DataStructs.ConvertToNumpyArray(fp, arr)
		  np_act_fps.append(arr)

		np_act_fps = numpy.array(np_act_fps)
		# print np_act_fps
		# print np_act_fps.shape
		act_lables = np.ones(np_act_fps.shape[0])

		np_dec_fps = []
		for fp in dec_fps:
		  arr = numpy.zeros((1,))
		  DataStructs.ConvertToNumpyArray(fp, arr)
		  np_dec_fps.append(arr)

		np_dec_fps = numpy.array(np_dec_fps)
		dec_lables = np.zeros(np_dec_fps.shape[0])

		X = numpy.concatenate((np_act_fps,np_dec_fps),axis=0)
		labels = numpy.concatenate((act_lables,dec_lables))
		W = metrics.silhouette_samples(X, labels, metric='euclidean')
		# print pro_name
		# print W.shape
		# print W[0:30]
		# print numpy.mean(W[0:30])
		# S = metrics.silhouette_score(X, labels, metric='euclidean')
		# print pro_name
		# print S
		silhouette_dict[pro_name]=numpy.mean(W[0:30])
	return silhouette_dict




def save(path, ext='png', close=True, verbose=True):
	
	# Extract the directory and filename from the given path
	directory = os.path.split(path)[0]
	filename = "%s.%s" % (os.path.split(path)[1], ext)
	if directory == '':
		directory = '.'
 
	# If the directory does not exist, create it
	if not os.path.exists(directory):
		os.makedirs(directory)
 
	# The final path to save to
	savepath = os.path.join(directory, filename)
 
	if verbose:
		print("Saving figure to '%s'..." % savepath),
 
	# Actually save the figure
	# plt.savefig(savepath)
	plt.savefig(savepath+'.eps', format='eps',dpi=900)#bbox_extra_artists=(lgd,), bbox_inches='tight'
	if close:
		plt.close()
 
	if verbose:
		print("Done")


def get_PF_Scores():
	PF_dict={}
	FF_tar_dict = {'2y6o_1N1':'689','1au8_0H8':'832','1yow_P0E':'692','3v2y_ML5':'466','3poo_S69':'548','5exm_5ST':'846','4xe4_NAG':'852','5cxv_0HK':'859','1yow_P0E':'600','5tn7_7G2':'713'}

	DUDE_MUV_file=open('DUDE_MUV.out')
	DUDE_MUV_list = list(DUDE_MUV_file)
	MUV_DUDE_dict={}
	
	for line in DUDE_MUV_list:
		ele=line.split()
		MUV=ele[0]
		DUDE=ele[1]
		score = -1*float(ele[-1])
		if MUV not in MUV_DUDE_dict.keys():
			MUV_DUDE_dict[MUV]=[]
		MUV_DUDE_dict[MUV].append(score)

	print ("MUV_DUDE_dict")
	print (MUV_DUDE_dict.keys())

	for MUV in MUV_DUDE_dict.keys():
		
		S = MUV_DUDE_dict[MUV]
		#s = numpy.mean(S)
		s = numpy.max(S)
		PF_dict[FF_tar_dict[MUV]]=s

	#PF_dict['600']=PF_dict['692']
	return PF_dict


def plot_data():
	#silhouette_dict=get_silhouette()
	PF_dict=get_PF_Scores()
	print (PF_dict)

	print PF_dict


	silhouette_dict={'846':0.0696665713558, '600':-0.00947942092601, '692':0.00677807984532, '859':0.00213644924873, '852':0.0536050806794, '548':0.049013577597, '713':-0.00076905984951, '733':-0.00283923942598, '466':-0.0101189101208, '689':0.000810431269303, '832':0.0306981850825}

	#mean
	# results={
	# '846':(0.0696665713558,1.1162222222222222),
	# '600':(-0.00947942092601,1.3975757575757575),
	# '692':(0.00677807984532,1.3975757575757575),
	# '859':(0.00213644924873,0.77167676767676752),
	# '852':(0.0536050806794,0.40196969696969698),
	# '548':(0.049013577597,1.871171717171717),
	# '713':(-0.00076905984951,1.0085050505050503),
	# '466':(-0.0101189101208,1.174070707070707),
	# '689':(0.000810431269303,1.8117575757575759), 
	# '832':(0.0306981850825,1.2264646464646465),
	# }

	#max
	results={
	'846':(0.0696665713558,4.935),
	'600':(-0.00947942092601,4.739),
	'692':(0.00677807984532,4.739),
	'859':(0.00213644924873,3.85),
	'852':(0.0536050806794,0.807),
	'548':(0.049013577597,5.91),
	'713':(-0.00076905984951,7.183),
	'466':(-0.0101189101208,3.877),
	'689':(0.000810431269303,5.835), 
	'832':(0.0306981850825,4.875),
	}

	X=[]
	Y=[]
	L=[]
	for k in results.keys():
		print (k)
		x=results[k][0]
		y=results[k][1]
		l=k
		X.append(x)
		Y.append(y)
		L.append(l)


	fig, ax = plt.subplots()
	ax.scatter(X, Y)
	for i in range (len(X)):
		ax.annotate(L[i], (X[i], Y[i]),fontsize=8)


	
	save('../silhouette_PF_051619', ext="png", close=True, verbose=True)




			
plot_data()