implemented --fine_plot option

Treemmer.py

@@ -10,8 +10,6 @@ import argparse
 
 
 
-	
-
 ############################################################			define arg type float 0 < X > 1		###############################################################
 
 def restricted_float(x):
@@ -221,7 +219,7 @@ def write_stop(t,output1,output2):
 
 
 
-######   SOFTWARE START
+######   SOFTWARE STARTS
 
 
 
@@ -230,14 +228,15 @@ def write_stop(t,output1,output2):
 parser = argparse.ArgumentParser()
 
 parser.add_argument('INFILE',type=str,help='path to the newick tree')
-parser.add_argument('-r','--resolution', metavar='INT', default=1,help='number of leaves top prune at each iteration (default: 1)',type =int, nargs='?')
+parser.add_argument('-X','--stop_at_X_leaves', metavar='0-n_leaves', default='0', help='stop pruning when the number of leaves =  X', type =int, nargs='?')
+parser.add_argument('-RTL','--stop_at_RTL', metavar='0-1', default='0', help='stop pruning when the relative tree length falls below RTL', type =restricted_float,nargs='?')
+parser.add_argument('-r','--resolution', metavar='INT', default=1,help='number of leaves to prune at each iteration (default: 1)',type =int, nargs='?')
+parser.add_argument('-p','--solve_polytomies',help='resolve polytomies at random (default: FALSE)',action='store_true',default =False)
+parser.add_argument('-lp','--leaves_pair', metavar='0,1,2', default=2,help='After the pair of leaves with the smallest distance is dentified Treemmer prunes: 0: the longest leaf\n1: the shortest leaf\n2: random choice (default)',type =int, nargs='?')
+parser.add_argument('-np','--no_plot',help='do not load matplotlib and plot (default: FALSE)',action='store_true',default =False)
+parser.add_argument('-fp','--fine_plot',help='when --resolution > 1, plot RTL vs n leaves every time a leaf is pruned  (default: FALSE => plot every X leaves (X = -r))',action='store_true',default =False)
 parser.add_argument('-c','--cpu', metavar='INT', default=1,help='number of cpu to use (default: 1)',type =int, nargs='?')
 parser.add_argument('-v' ,'--verbose', metavar='0,1,2', default='0', help='0: silent, 1: show progress, 2: print tree at each iteration, 3: only for testing (findN), 4: only for testing (prune_t) (default: 1)', type =int, nargs='?',choices=[0,1,2,3,4])
-parser.add_argument('-p','--solve_polytomies',help='resolve polytmies at random (default: FALSE)',action='store_true',default =False)
-parser.add_argument('-np','--no_plot',help='do not load matplotlib and plot (default: FALSE)',action='store_true',default =False)
-parser.add_argument('-X','--stop_at_X_leaves', metavar='0-n_leaves', default='0', help='stop pruning when the number of leaves fall below  X (integer)', type =int, nargs='?')
-parser.add_argument('-RTL','--stop_at_RTL', metavar='0-1', default='0', help='stop pruning when the relative tree length falls below RTL (decimal number between 0 and 1)', type =restricted_float,nargs='?')
-parser.add_argument('-lp','--leaves_pair', metavar='0,1,2', default=2,help='After the pair of leaves with the smallest distance is dentified Treemmer prunes: 0: the longest leaf\n1: the shortest leaf\n2: random choice (default)',type =int, nargs='?')
 
 
 arguments = parser.parse_args()
@@ -294,7 +293,7 @@ while (len(t) > 3):
 	DLIST = Parallel(n_jobs=arguments.cpu)(delayed(parallel_loop)(i) for i in range(0,arguments.cpu))
 	result = {}
 
-	for d in DLIST:			#when running in parallel DLIST is updated in a weird way, it is a dict of dicts, this for loop merge them all in one
+	for d in DLIST:								#when running in parallel DLIST is updated in a weird way, it is a dict of dicts, this for loop merge them all in one
 		result.update(d)
 
 	DLIST=result
@@ -314,8 +313,15 @@ while (len(t) > 3):
 		TL=TL-dist
 		DLIST=prune_dist_matrix(DLIST,leaf_to_p)
 		rel_TL=TL/TOT_TL
+		
+
+		if (arguments.fine_plot):										# plot point in rtld after every leaf independently of -r 
+			output.append (str(rel_TL) + '	' + str(len(t)))
+			length=len(t)	
+			x.append(length)
+			y.append(rel_TL)
 
-		if arguments.stop_at_X_leaves:
+		if arguments.stop_at_X_leaves:										# if stop criterium is met (X) ==> output
 			if arguments.stop_at_X_leaves >= len(t):
 				output1=arguments.INFILE+"_trimmed_tree_X_" + str(arguments.stop_at_X_leaves)
 				output2=arguments.INFILE+"_trimmed_list_X_" + str(arguments.stop_at_X_leaves)
@@ -323,8 +329,8 @@ while (len(t) > 3):
 				stop=1
 				break
 			
-		if arguments.stop_at_RTL:
-			if arguments.stop_at_RTL >= rel_TL:
+		if arguments.stop_at_RTL:										# if stop criterium is met (RTL) ==> output
+			if arguments.stop_at_RTL >= rel_T:
 				output1=arguments.INFILE+"_trimmed_tree_RTL_" + str(arguments.stop_at_RTL)
 				output2=arguments.INFILE+"_trimmed_list_RTL_" + str(arguments.stop_at_RTL) 
 				write_stop(t,output1,output2)
@@ -344,10 +350,11 @@ while (len(t) > 3):
 	if (stop ==1):
 		break	
 	
-	output.append (str(rel_TL) + '	' + str(len(t)))
-	length=len(t)	
-	x.append(length)
-	y.append(rel_TL)
+	if not (arguments.fine_plot):
+		output.append (str(rel_TL) + '	' + str(len(t)))
+		length=len(t)	
+		x.append(length)
+		y.append(rel_TL)
 		
 
 	if arguments.verbose==1:
@@ -372,69 +379,9 @@ if not arguments.no_plot:
 	plt.ylim(0,1.1)
 	plt.xlabel('Number of leaves')
 	plt.ylabel('Relative tree length')
-	plt.savefig(arguments.INFILE+'_TLD.png')
+	plt.savefig(arguments.INFILE+'_res_'+ str(arguments.resolution)+'_TLD.png')
 			
 
 
 
 
-
-
-
-########################################### LOOP THRU ALL LEAVES (replaced by joblib parallel line  #############################################
-
-
-#	for leaf in leaves:
-#		counter1 = counter1 + 1
-#		print "iter find N : "+ str(counter1)
-#		N_list=find_N(t,leaf)
-#		print N_list
-#		if N_list:
-#			DLIST.update(N_list)
-
-
-
-
-###########################################		CALCULATE MATRIX DISTANCE into simple DICTIONARY	very slow!! dendropy is orders of magnitude faster #############################################
-
-#def make_dist_matrix(leaves):
-#	count=0
-#	dlist = {}
-#	for x in range (0,len(leaves)-1):
-#		count=count+1
-#
-#		for y in range (x+1,len(leaves)):
-#
-#			DIS = leaves[x].get_distance(leaves[y])
-#			dlist.update({leaves[x].name + "," +leaves[y].name : DIS})
-#	
-#	return dlist
-
-#print dlist
-
-
-###########################################		CALCULATE MATRIX DISTANCE into nested DICTIONARY			##############################################
-
-
-#m = {}
-#for x in range (0,len(leaves)-2):
-#	d = {}
-#	for y in range (x+1,len(leaves)-1):
-#
-#		DIS = leaves[x].get_distance(leaves[y])
-#		#print str(leaves[x]) + str(leaves[y]) + "\n" + str(DIS)
-#		d.update({leaves[y].name : DIS})
-	#print "D"#
-	#print d
-#	m.update({leaves[x].name : d})
-	#print "M"
-	#print m
-
-#print m
-
-
-
-
-
-
-