-
Notifications
You must be signed in to change notification settings - Fork 2
/
protein_coexpression.py
308 lines (228 loc) · 10.8 KB
/
protein_coexpression.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
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
import matplotlib.pyplot as plt
import numpy as np
import os
import pandas as pd
from scipy.cluster.hierarchy import fcluster
from lifelines import KaplanMeierFitter
from lifelines.statistics import logrank_test
import seaborn
#*-------------------------------
#Recurrence
#Path to the clinical data about recurrence.
clinical_path = "rawdata/clinical_data.csv"
clinical_df = pd.read_csv(clinical_path, index_col=["ID"])
#Path to interaction matrices
matrices_path = "intermediate_data/coexpression_matrices/"
# Out of all proteins, only include the functional proteins
markers_to_include = [2,6,14,16,22,23,24,25,26,27,28,29,30,31,35,37,38,39]
columns = []
feature_list = []
#Iterate over all of the co-occurrence matrices in this certain radius.
for patient_index in range(len(os.listdir(matrices_path))):
patient_glcm = os.listdir(matrices_path)[patient_index]
#Skip over the pesky .DS_Store file that shows up in Mac file systems.
if (patient_glcm[0] == "."):
continue
#Find the internal_ID of the current patient.
identifier = patient_glcm.split(".")[0]
#Read the co-occurrence matrix of the current patient and cast to a numpy array excluding the first col.
current_patient_glcm = pd.read_csv(os.path.join(matrices_path, patient_glcm), index_col="Unnamed: 0")
current_patient_glcm.set_index(current_patient_glcm.columns, inplace=True)
#feature_name = current_patient_glcm.columns[chosen_feature]
np_glcm = current_patient_glcm.to_numpy()
patient_features = []
#Flatten the co-occurrence matrix into a feature vector.
#Cannot simply run np.flatten because the matrix is symmetrical - I only want the top-right triangle
#But, we don't want to include the diagonal
for row in range(np_glcm.shape[0]):
for column in range(row+1, np_glcm.shape[1]):
if row in markers_to_include and column in markers_to_include:
if patient_index == 0:
feature_name = current_patient_glcm.index[row] + " " + current_patient_glcm.columns[column]
columns.append(feature_name)
patient_features.append(np_glcm[row][column])
#Find the recurrence outcome of this current patient and how long it took for them to recur
#Add these values to the feature list for eventual use in the Kaplan-Meier plot.
try:
patient_features.append(clinical_df.at[int(identifier), "Recurrence"])
except KeyError:
continue
patient_features.append(clinical_df.at[int(identifier), "Recurrence_time"])
patient_features.append(int(identifier))
feature_list.append(patient_features)
#Determine the names of the columns in the DataFrame for easier future access.
columns.append("Recurrence")
columns.append("Recurrence_time")
columns.append("ID")
#Create a dataframe using the features, the recurrence events, and the time taken to recur.
features_df = pd.DataFrame(feature_list, columns=columns)
#Obtain a versino of this dataframe with only the features.
data_only = features_df.drop(columns=["Recurrence_time", "Recurrence", "ID"])
data_only["Duplicate"] = features_df[feature_name]
data_only.set_index(features_df["ID"], inplace=True)
#Create the dendrogram.
clustergram = seaborn.clustermap(data_only, method="weighted",
metric="braycurtis", standard_scale = 1, cmap="viridis", figsize=(10,8), cbar_pos=None)
plt.close()
#Number of clusters to take from the dendrogram
k = 2
#Use scipy fcluster to find clusters from the clustered dendrograms.
clusters = list(fcluster(clustergram.dendrogram_row.linkage, k, criterion='maxclust'))
unique_clusters = len(np.unique(np.array(clusters)))
if (unique_clusters < 2):
print ("Not able to find more than one cluster. ") #If there was no clustering, this method failed.
#Create a new column in the DataFrame that includes what cluster each patient falls into
features_df["clust"] = clusters
first_cluster_count = clusters.count(1)
second_cluster_count = clusters.count(2)
#Define the KaplanMeierFitter
kmf = KaplanMeierFitter()
#T = Time, E=Event. These are the two parameters that go into Kaplan-Meier curves.
T = features_df["Recurrence_time"]
E = features_df["Recurrence"]
group1 = (features_df["clust"] == 1)
group2 = (features_df["clust"] == 2)
T1 = T[group1]
E1 = E[group1]
T2 = T[group2]
E2 = E[group2]
color_clust1 = "#F39B7FFF"
color_clust2 = "#4DBBD5FF"
# Just for visualization purposes, make the worse-outcome cluster orange and the other blue
if E1.mean() < E2.mean():
T1 = T[group2]
E1 = E[group2]
T2 = T[group1]
E2 = E[group1]
features_df["clust"][group1] = 2
features_df["clust"][group2] = 1
first_cluster_count = len(T1.index)
second_cluster_count = len(T2.index)
results_first = logrank_test(T1,T2, event_observed_A=E1, event_observed_B=E2)
p1 = round(results_first.p_value, 4)
plt.figure(figsize=(10,8))
kmf.fit(T1, E1, label='Cluster 1: n=' + str(first_cluster_count))
ax = kmf.plot(ci_show=False, show_censors=True, color=color_clust1, lw=5)
kmf.fit(T2, E2, label='Cluster 2: n=' + str(second_cluster_count))
ax = kmf.plot(ax=ax, ci_show=False, show_censors=True, color=color_clust2, lw=5)
plt.annotate("Log-rank p: " + str(p1), xy=(0.6, 0.15), xycoords="figure fraction", fontsize=25,
bbox=dict(facecolor='none', edgecolor='black', alpha=0.3, boxstyle="Round, pad=0.5, rounding_size=0.2"))
plt.legend(fontsize=25)
plt.xlabel("Time (days)", fontsize=25)
plt.xticks(fontsize=15)
plt.ylabel("Proportion Alive", fontsize=25)
plt.yticks(fontsize=15)
plt.ylim(0,1)
plt.tight_layout()
plt.savefig("results/coexpression_km_recurrence.png", dpi=300)
plt.close()
#*-------------------------------
#Survival
#Path to the clinical data about recurrence.
clinical_path = "rawdata/clinical_data.csv"
clinical_df = pd.read_csv(clinical_path, index_col=["ID"])
#Path to interaction matrices
matrices_path = "intermediate_data/coexpression_matrices/"
# Out of all proteins, only include the functional proteins
markers_to_include = [2,6,14,16,22,23,24,25,26,27,28,29,30,31,35,37,38,39]
columns = []
feature_list = []
#Iterate over all of the co-occurrence matrices in this certain radius.
for patient_index in range(len(os.listdir(matrices_path))):
patient_glcm = os.listdir(matrices_path)[patient_index]
#Skip over the pesky .DS_Store file that shows up in Mac file systems.
if (patient_glcm[0] == "."):
continue
#Find the internal_ID of the current patient.
identifier = patient_glcm.split(".")[0]
#Read the co-occurrence matrix of the current patient and cast to a numpy array excluding the first col.
current_patient_glcm = pd.read_csv(os.path.join(matrices_path, patient_glcm), index_col="Unnamed: 0")
current_patient_glcm.set_index(current_patient_glcm.columns, inplace=True)
#feature_name = current_patient_glcm.columns[chosen_feature]
np_glcm = current_patient_glcm.to_numpy()
patient_features = []
#Flatten the co-occurrence matrix into a feature vector.
#Cannot simply run np.flatten because the matrix is symmetrical - I only want the top-right triangle
#But, have to include the diagonal as well.
for row in range(np_glcm.shape[0]):
for column in range(row+1, np_glcm.shape[1]):
if row in markers_to_include and column in markers_to_include:
if patient_index == 0:
feature_name = current_patient_glcm.index[row] + " " + current_patient_glcm.columns[column]
columns.append(feature_name)
patient_features.append(np_glcm[row][column])
#Find the recurrence outcome of this current patient and how long it took for them to recur
#Add these values to the feature list for eventual use in the Kaplan-Meier plot.
try:
patient_features.append(clinical_df.at[int(identifier), "Survival"])
except KeyError:
continue
patient_features.append(clinical_df.at[int(identifier), "Survival_time"])
patient_features.append(int(identifier))
feature_list.append(patient_features)
#Determine the names of the columns in the DataFrame for easier future access.
columns.append("Survival")
columns.append("Survival_time")
columns.append("ID")
#Create a dataframe using the features, the recurrence events, and the time taken to recur.
features_df = pd.DataFrame(feature_list, columns=columns)
#Obtain a versino of this dataframe with only the features.
data_only = features_df.drop(columns=["Survival", "Survival_time", "ID"])
data_only["Duplicate"] = features_df[feature_name]
data_only.set_index(features_df["ID"], inplace=True)
#Create the dendrogram.
clustergram = seaborn.clustermap(data_only, method="weighted",
metric="cosine", standard_scale = 1, cmap="viridis", figsize=(10,8), cbar_pos=None)
plt.close()
#Number of clusters to take from the dendrogram
k = 2
#Use scipy fcluster to find clusters from the clustered dendrograms.
clusters = list(fcluster(clustergram.dendrogram_row.linkage, k, criterion='maxclust'))
unique_clusters = len(np.unique(np.array(clusters)))
if (unique_clusters < 2):
print ("Not able to find more than one cluster. ") #If there was no clustering, this method failed.
#Create a new column in the DataFrame that includes what cluster each patient falls into
features_df["clust"] = clusters
first_cluster_count = clusters.count(1)
second_cluster_count = clusters.count(2)
#Define the KaplanMeierFitter
kmf = KaplanMeierFitter()
#T = Time, E=Event. These are the two parameters that go into Kaplan-Meier curves.
T = features_df["Survival_time"]
E = features_df["Survival"]
group1 = (features_df["clust"] == 1)
group2 = (features_df["clust"] == 2)
T1 = T[group1]
E1 = E[group1]
T2 = T[group2]
E2 = E[group2]
color_clust1 = "#F39B7FFF"
color_clust2 = "#4DBBD5FF"
# Just for visualization purposes, make the worse-outcome cluster orange and the other blue
if E1.mean() < E2.mean():
T1 = T[group2]
E1 = E[group2]
T2 = T[group1]
E2 = E[group1]
features_df["clust"][group1] = 2
features_df["clust"][group2] = 1
first_cluster_count = len(T1.index)
second_cluster_count = len(T2.index)
results_first = logrank_test(T1,T2, event_observed_A=E1, event_observed_B=E2)
p1 = round(results_first.p_value, 4)
plt.figure(figsize=(10,8))
kmf.fit(T1, E1, label='Cluster 1: n=' + str(first_cluster_count))
ax = kmf.plot(ci_show=False, show_censors=True, color=color_clust1, lw=5)
kmf.fit(T2, E2, label='Cluster 2: n=' + str(second_cluster_count))
ax = kmf.plot(ax=ax, ci_show=False, show_censors=True, color=color_clust2, lw=5)
plt.annotate("Log-rank p: " + str(p1), xy=(0.6, 0.15), xycoords="figure fraction", fontsize=25,
bbox=dict(facecolor='none', edgecolor='black', alpha=0.3, boxstyle="Round, pad=0.5, rounding_size=0.2"))
plt.legend(fontsize=25)
plt.xlabel("Time (days)", fontsize=25)
plt.xticks(fontsize=15)
plt.ylabel("Proportion Alive", fontsize=25)
plt.yticks(fontsize=15)
plt.ylim(0,1)
plt.tight_layout()
plt.savefig("results/coexpression_km_survival.png", dpi=300)
plt.close()