-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathIntro.py
286 lines (261 loc) · 12.7 KB
/
Intro.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
# -*- coding: utf-8 -*-
"""
Created on Fri Sep 28 18:44:11 2018
@author: Admin
"""
import numpy as np
import networkx as nx
import matplotlib.pyplot as plt
import pandas as pd
from scipy import optimize
#%%
def ldata(archive):
f=open(archive)
data=[]
for line in f:
line=line.strip()
col=line.split()
data.append(col)
return data
#%%
prot = ldata('C:/Users/Admin/Documents/GitHub/Redes_TP2/data/yeast_AP-MS.txt')
bina = ldata('C:/Users/Admin/Documents/GitHub/Redes_TP2/data/yeast_Y2H.txt')
lit = ldata('C:/Users/Admin/Documents/GitHub/Redes_TP2/data/yeast_LIT.txt')
litreg = ldata('C:/Users/Admin/Documents/GitHub/Redes_TP2/data/yeast_LIT_Reguly.txt')
essential = ldata('C:/Users/Admin/Documents/GitHub/Redes_TP2/data/Essential_ORFs_paperHe.txt')
#%% LIMPIO LOS DATOS
ess = [a[1] for a in essential[2:1158]]
litr = [a[0:2] for a in litreg[1:]]
#%%
P = nx.Graph()
P.add_edges_from(prot)
B = nx.Graph()
B.add_edges_from(bina)
L = nx.Graph()
L.add_edges_from(lit)
LR = nx.Graph()
LR.add_edges_from(litr)
#%%
def directed(A):
for i in range(0,len(A)):
for r in range(i+1,len(A)):
if A[i][0] == A[r][1] and A[i][1] == A[r][0]:
return "SI"
return "NO"
def topologia(redes, edges): #dar vector con redes ordenadas y sus bases de datos
R = redes
L = edges
N = np.empty_like(R) # número de nodos de la red
E = np.empty_like(R) # número de enlaces de la red
R_mean = np.empty_like(R) # grado medio de la red
R_max = np.empty_like(R) # grado máximo
R_min = np.empty_like(R) # grado mínimo
dens = np.empty_like(R) # densidad de la red
clu = np.empty_like(R) # coeficiente de clustering local
clu_d = np.empty_like(R) # coeficiente de clustering global/transitividad
diam = np.empty_like(R) # diámetro de la red
for i in range(len(R)):
N[i] = R[i].number_of_nodes()
E[i] = R[i].number_of_edges()
R_mean[i] = np.mean([a[1] for a in list(R[i].degree())])
R_max[i] = np.max([a[1] for a in list(R[i].degree())])
R_min[i] = np.min([a[1] for a in list(R[i].degree())])
dens[i] = nx.density(R[i])
clu[i] = nx.average_clustering(R[i])
clu_d[i] = nx.transitivity(R[i])
diam[i] = nx.diameter(max(nx.connected_component_subgraphs(R[i]), key=len))
dirigido = np.empty_like(R)
for i in range(len(dirigido)):
dirigido[i] = directed(L[i])
return N, E, R_mean, R_max, R_min, dens, clu, clu_d, diam, dirigido
#%%
A = topologia([P,B,L,LR], [prot, bina, lit, litr])
#%%
tabla = pd.DataFrame({"Red":["Proteicas", "Binarias","Literatura", "Literatura Regulada"],"# de nodos":A[0],"# total de enlaces":A[1],"Grado medio":A[2],"Coef. de Clust. red": A[6],"Dirigida?":A[9]})
print(tabla)
#%% #función para asignar si un nodo es esenecial o no para todas las redes al mismo tiempo
def daresencialidad(redes, ess):
R = redes
esencial = []#np.empty_like((len(R), 1)) ##vamos a tener que usar un diccionario porque hay distinto numero de nodos
for i in range(len(R)):
D = dict()
a = list(R[i].nodes())
for j in range(len(a)):
D[j] = 0
for l in range(len(ess)):
if a[j] == ess[l]:
D[j] = 1
break
esencial.append(D)
return esencial
#%% #meto el atributo esencialidad en los nodos de cada red
def atribuir(redes, ess):
R = redes
for i in range(len(R)):
for n,e,g in zip(R[i].nodes, daresencialidad(redes, ess)[i].values(), [a[1] for a in list(R[i].degree())]):
R[i].nodes[n]['Esencialidad'] = e
R[i].nodes[n]['Grado'] = g
#%%
atribuir([P,B,L,LR],ess)
#%%
def eshub(redesconatributos, porcentajes): #np.linspace(0,1,11)
ess_percent = [] #aca voy a appendear los vectores para cada red, osea que va a tener len = len(R)
R = redesconatributos
for i in range(len(R)):
a = []
for j,val in enumerate(porcentajes):
cant_nodos_esenciales = np.sum([b[1]['Esencialidad'] for b in list(sorted(R[i].nodes.data(),key = lambda x: -x[1]['Grado']))[0:int(val*R[i].number_of_nodes())]])
a.append(cant_nodos_esenciales/int(val*R[i].number_of_nodes())) #guardo el porcentaje de nodos esenciales hasta el número de nodos considerados como hubs por vez
ess_percent.append(a)
return ess_percent
#%%
redesconatributos = [P,B,L,LR]
porcentajes = np.concatenate([np.linspace(0.001,0.1,50), np.linspace(0.1,1,100)[1:]])
ess_percent = eshub(redesconatributos, porcentajes)
redes = ["Proteicas", "Binarias","Literatura", "Literatura Reguly"]
#%%
#ess percent va a ser un vector con tantos vectores como redes
plt.figure(1)
plt.plot(porcentajes, ess_percent[0],'.-', label = '%s'%redes[0])
plt.plot(porcentajes, ess_percent[1], '.-', label = '%s'%redes[1])
plt.plot(porcentajes, ess_percent[2], '.-',label = '%s'%redes[2])
plt.plot(porcentajes, ess_percent[3],'.-', label = '%s'%redes[3])
plt.ylim((0.1,1.1))
plt.title('Relación Grado-Esencialidad')
plt.xlabel('Hub cut-off')
plt.ylabel('Porcentaje de hubs esenciales')
plt.legend()
plt.grid(True)
#%%
#def centralidades(redesND, redD):
# R = redesND
# R_GC = np.empty_like(R)
# T = redD
#
#
# for i in range(len(R_GC)):
# R_GC[i] = max(nx.connected_component_subgraphs(R[i]), key=len)
# for n,deg, eigen, between, current in zip(R_GC[i].nodes, list(nx.degree_centrality(R_GC[i]).values()), list(nx.eigenvector_centrality(R_GC[i]).values()), list(nx.betweenness_centrality(R_GC[i]).values()), list(nx.current_flow_betweenness_centrality(R_GC[i]).values())):
# R_GC[i].nodes[n]['degree'] = deg
# R_GC[i].nodes[n]['eigenvector'] = eigen
# R_GC[i].nodes[n]['betweenness'] = between
# R_GC[i].nodes[n]['current'] = current
#
# for j in range(len(T)):
# T_GC = max(nx.connected_component_subgraphs(T), key=len)
# for n,deg, eigen, between, current, degin, degout in zip(T_GC.nodes, list(nx.degree_centrality(T_GC).values()), list(nx.eigenvector_centrality(T_GC).values()), list(nx.betweenness_centrality(T_GC).values()), list(nx.current_flow_betweenness_centrality(T_GC).values()), list(nx.in_degree_centrality(T_GC).values()), list(nx.out_degree_centrality(T_GC).values())):
# T_GC[j].nodes[n]['degree'] = deg
# T_GC[j].nodes[n]['eigenvector'] = eigen
# T_GC[j].nodes[n]['betweenness'] = between
# T_GC[j].nodes[n]['current'] = current
# T_GC[j].nodes[n]['degree_in'] = degin
# T_GC[j].nodes[n]['degree_out'] = degout
#%%
redes_analisis = [P,B,L,LR.to_undirected()]
def centralidades2(redes):
R = redes
R_GC = np.empty_like(R)
for i in range(len(R_GC)):
R_GC[i] = max(nx.connected_component_subgraphs(R[i]), key=len)
for n,deg, eigen, between, current in zip(R_GC[i].nodes, list(nx.degree_centrality(R_GC[i]).values()), list(nx.eigenvector_centrality(R_GC[i]).values()), list(nx.betweenness_centrality(R_GC[i]).values()), list(nx.current_flow_betweenness_centrality(R_GC[i]).values())):
R_GC[i].nodes[n]['degree'] = deg
R_GC[i].nodes[n]['eigenvector'] = eigen
R_GC[i].nodes[n]['betweenness'] = between
R_GC[i].nodes[n]['current'] = current
return R_GC
#%%
R_GC = centralidades2(redes_analisis)
#%%
nodes = np.empty((4,len(R_GC)), dtype= object) #aca se van a guardar los nombres de los nodos a eliminar en cada caso(para cada centralidad)
for i in range(len(R_GC)):
largo = max(nx.connected_component_subgraphs(R_GC[i]), key=len).number_of_nodes()#int(np.sum([c[1]['Esencialidad'] for c in list(R_GC[i].nodes.data())]))
nodes[0,i] = [b[0] for b in list(sorted(R_GC[i].nodes.data(), key = lambda x: -x[1]['degree']))][0:largo]
nodes[1,i] = [b[0] for b in list(sorted(R_GC[i].nodes.data(), key = lambda x: -x[1]['eigenvector']))][0:largo]
nodes[2,i] = [b[0] for b in list(sorted(R_GC[i].nodes.data(), key = lambda x: -x[1]['betweenness']))][0:largo]
nodes[3,i] = [b[0] for b in list(sorted(R_GC[i].nodes.data(), key = lambda x: -x[1]['current']))][0:largo]
#%%
percent = np.linspace(0,1,10)
tamaños = np.empty((len(R_GC),len(percent),4),dtype= object) #aca voy a poner el tamaño de la componente gigante del grafo en cada caso, con los valores de cada red en columnas
#osea que la primera columna tendra el tamaño de la componente gigante habiendo sacado nada, 005% etc...
for i in range(len(R_GC)):
tamaños[i,0,:] = max(nx.connected_component_subgraphs(R_GC[i]), key=len).number_of_nodes() #innecesario porque R_GC ya es la componente gigante
for n, val in enumerate(percent[1:]):
for j in range(4):
red = R_GC[i].copy()
red.remove_nodes_from(nodes[j,i][0:int(val*len(nodes[j,i]))])
tamaños[i,n+1,j] = int(max(nx.connected_component_subgraphs(red), key=len).number_of_nodes())
#%%
nodos_esenciales = []
tamaño = []
for i in range(len(R_GC)):
red = R_GC[i].copy()
red_random = R_GC[i].copy()
nodos_esenciales.append([b[0] for b in list(sorted(R_GC[i].nodes.data(), key = lambda x: -x[1]['Esencialidad']))][0:np.sum([c[1]['Esencialidad'] for c in list(R_GC[i].nodes.data())])])
nodos_random.append()
red.remove_nodes_from(nodos_esenciales[i])
tamaño.append(int(max(nx.connected_component_subgraphs(red), key=len).number_of_nodes()))
red_random.remove_nodes_from()
tamaño_random.append()
#%%
from random import shuffle
nodos_random = np.empty((len(R_GC),len(percent)), dtype = object)
tamaño_random = np.empty((len(R_GC),len(percent)), dtype = object)
for i in range(len(R_GC)):
tamaño_random[i,0] = R_GC[i].number_of_nodes()
for n, val in enumerate(percent[1:]):
red_random = R_GC[i].copy()
nodos_random = list(R_GC[i].nodes())[0:int(val*R_GC[i].number_of_nodes())]
shuffle(nodos_random)
red_random.remove_nodes_from(nodos_random)
tamaño_random[i,n+1] = int(max(nx.connected_component_subgraphs(red_random), key=len).number_of_nodes())
#%% Grafico del efecto de eliminar nodos con distintos valores de conectividad
plt.figure(1, figsize=(15,10))
plt.subplot(221)
plt.title('Protéicas')
plt.xlabel('Fracción de nodos')
plt.ylabel('Fracción de componente gigante')
plt.plot(percent, tamaños[0,:,0]/tamaños[0,0,0], label = 'Grado')
plt.plot(percent, tamaños[0,:,1]/tamaños[0,0,1], label = 'Autovector')
plt.plot(percent, tamaños[0,:,2]/tamaños[0,0,2], label = 'Betweenness')
plt.plot(percent, tamaños[0,:,3]/tamaños[0,0,3], label = 'Corriente')
plt.plot(percent, tamaño_random[0,:]/tamaño_random[0,0], label = 'Azar')
plt.plot(np.sum([c[1]['Esencialidad'] for c in list(R_GC[0].nodes.data())])/R_GC[0].number_of_nodes(),tamaño[0]/R_GC[0].number_of_nodes(), 'd')
plt.grid(True)
plt.legend()
plt.subplot(222)
plt.title('Binarias')
plt.xlabel('Fracción de nodos')
plt.ylabel('Fracción de componente gigante')
plt.plot(percent, tamaños[1,:,0]/tamaños[1,0,0], label = 'Grado')
plt.plot(percent, tamaños[1,:,1]/tamaños[1,0,1], label = 'Autovector')
plt.plot(percent, tamaños[1,:,2]/tamaños[1,0,2], label = 'Betweenness')
plt.plot(percent, tamaños[1,:,3]/tamaños[1,0,3], label = 'Corriente')
plt.plot(percent, tamaño_random[1,:]/tamaño_random[1,0], label = 'Azar')
plt.plot(np.sum([c[1]['Esencialidad'] for c in list(R_GC[1].nodes.data())])/R_GC[1].number_of_nodes(),tamaño[1]/R_GC[1].number_of_nodes(), 'd')
plt.grid(True)
plt.legend()
plt.subplot(223)
plt.title('Literatura')
plt.xlabel('Fracción de nodos')
plt.ylabel('Fracción de componente gigante')
plt.plot(percent, tamaños[2,:,0]/tamaños[2,0,0], label = 'Grado')
plt.plot(percent, tamaños[2,:,1]/tamaños[2,0,1], label = 'Autovector')
plt.plot(percent, tamaños[2,:,2]/tamaños[2,0,2], label = 'Betweenness')
plt.plot(percent, tamaños[2,:,3]/tamaños[2,0,3], label = 'Corriente')
plt.plot(percent, tamaño_random[1,:]/tamaño_random[1,0], label = 'Azar')
plt.plot(np.sum([c[1]['Esencialidad'] for c in list(R_GC[2].nodes.data())])/R_GC[2].number_of_nodes(),tamaño[2]/R_GC[2].number_of_nodes(), 'd')
plt.grid(True)
plt.legend()
plt.subplot(224)
plt.title('Literatura Reguly')
plt.xlabel('Fracción de nodos')
plt.ylabel('Fracción de componente gigante')
plt.plot(percent, tamaños[3,:,0]/tamaños[3,0,0], label = 'Grado')
plt.plot(percent, tamaños[3,:,1]/tamaños[3,0,1], label = 'Autovector')
plt.plot(percent, tamaños[3,:,2]/tamaños[3,0,2], label = 'Betweenness')
plt.plot(percent, tamaños[3,:,3]/tamaños[3,0,3], label = 'Corriente')
plt.plot(percent, tamaño_random[1,:]/tamaño_random[1,0], label = 'Azar')
plt.plot(np.sum([c[1]['Esencialidad'] for c in list(R_GC[3].nodes.data())])/R_GC[3].number_of_nodes(),tamaño[3]/R_GC[3].number_of_nodes(), 'd')
plt.grid(True)
plt.legend()
#%%