-
Notifications
You must be signed in to change notification settings - Fork 1
/
cluster_CDRs.py
219 lines (171 loc) · 7.85 KB
/
cluster_CDRs.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
# By Neerja Thakkar for "Balancing sensitivity and specificity in
# distinguishing TCR groups by CDR sequence similarity"
# See README for license information
from matplotlib import pyplot as plt
from scipy.cluster.hierarchy import dendrogram, linkage, fcluster
from Bio import motifs
from Bio.Seq import Seq
from Bio.Alphabet import IUPAC
from Bio.SeqIO.FastaIO import SimpleFastaParser
from sw_scoring import getDistanceSW
def construct_distance_vector(Xlabels, length_dep=True, gap_penalty=-10):
distances = []
for i in range(len(Xlabels)):
for j in range(i+1, len(Xlabels)):
dist = getDistanceSW(Xlabels[i], Xlabels[j], length_dep=length_dep, gap_penalty=gap_penalty)
distances.append(dist)
return distances
def clusters_to_dict(cluster_arr, Xlabels):
clusters = {}
for i in range(len(cluster_arr)):
if cluster_arr[i] in clusters:
clusters[cluster_arr[i]].append(Xlabels[i])
else:
clusters[cluster_arr[i]] = [Xlabels[i]]
return clusters
def cluster_data(Xlabels, max_distance=0.2, colors_dict=False, structural_cluster_dict=None, length_dep=True, gap_penalty=-10):
X = construct_distance_vector(Xlabels, length_dep, gap_penalty)
Z = linkage(X, 'average')
cluster_arr = fcluster(Z, max_distance, criterion='distance')
clusters = clusters_to_dict(cluster_arr, Xlabels)
return clusters
def cluster_data_and_generate_dendrogram(Xlabels, max_distance=0.2, colors_dict=False, structural_cluster_dict=None, length_dep=True, gap_penalty=-10):
X = construct_distance_vector(Xlabels, length_dep, gap_penalty)
Z = linkage(X, 'average')
for i in range(len(Xlabels)):
print(str(i) + ": " + Xlabels[i])
plt.figure(figsize=(10, 5))
plt.title('Hierarchical Clustering Dendrogram')
plt.xlabel('sample index')
plt.ylabel('distance')
dendrogram(
Z,
leaf_rotation=90., # rotates the x axis labels
leaf_font_size=8., # font size for the x axis labels
labels=Xlabels
)
if colors_dict:
num_to_color = {1: 'r', 2: 'g', 3: 'b', 4: 'm', 5: 'y', 6: 'c'}
ax = plt.gca()
x_labels = ax.get_xticklabels()
for x in x_labels:
if structural_cluster_dict[x.get_text()] != 0:
x.set_color(num_to_color[structural_cluster_dict[x.get_text()]])
plt.show()
cluster_arr = fcluster(Z, max_distance, criterion='distance')
clusters = clusters_to_dict(cluster_arr, Xlabels)
return clusters
def create_motif_from_fasta_file(fasta_filename, out_filename, generate_pssm=False):
instances = []
with open(fasta_filename) as in_handle:
for title, seq in SimpleFastaParser(in_handle):
instances.append(Seq(seq, IUPAC.protein))
m = motifs.create(instances, IUPAC.protein)
m.weblogo(out_filename,
show_xaxis=False,
show_yaxis=False,
show_errorbars=False,
unit_name='',
show_fineprint=False,
format='pdf')
if generate_pssm:
pssm_file = open(out_filename[:-10] + "_pssm.txt", "w+")
for i in range(len(m.pwm)):
for j in range(len(m.pwm[i])):
pssm_file.write(str(m.pwm[i][j]))
pssm_file.write("\t")
pssm_file.write("\n")
pssm_file.close()
# takes in the clusters
# generates the correctly formatted text file of the CDRs in the cluster
# if a size is given, only clusters of that size will be written to files
def generate_motifs(clusters, file_path, clust_size=None):
for key in clusters:
print ("processing cluster " + str(key))
process_cluster = True
if clust_size:
size = len(clusters[key])
if size != clust_size:
process_cluster = False
if process_cluster:
file_name = file_path + str(key) + ".txt"
f = open(file_name, "w")
clust = clusters[key]
i = 0
for read in clust:
f.write(">" + str(i) + "\n")
f.write(read + "\n\n")
i += 1
f.close()
create_motif_from_fasta_file(file_name, file_name[:-4] + "_motif.pdf")
def generate_motif_from_cluster(key, cluster, file_path, clust_size=None):
clusters = {key: cluster}
generate_motifs(clusters, file_path, clust_size)
# n = size of clusters in smallest distance clustering, to keep track of
# clusters2 = cluster_data(Xlabels, max_distance=0.2)
# clusters3 = cluster_data(Xlabels, max_distance=0.3)
# clusters4 = cluster_data(Xlabels, max_distance=0.4)
# default is to find the cluster progression of all of the clusters in clusters2
# can also pass in a cluster in clust, which will return the progression of a specific cluster
def cluster_progression(out_dir, clusters2, clusters3, clusters4, n=None, clust=None):
motif_names = []
if clust != None:
for c in clusters2:
if clust == clusters2[c]:
# process cluster - generate motif
motif_name = str(c) + "_0.2"
generate_motif_from_cluster(motif_name, clusters2[c], out_dir)
motif_names.append(out_dir + motif_name + "_pssm.txt")
# get first CDR - all unique, so can use to find how the cluster evolved
first_cdr = clusters2[c][0]
for j in clusters3:
if first_cdr in clusters3[j]:
motif_name = str(c) + "_0.3"
generate_motif_from_cluster(motif_name, clusters3[j], out_dir)
motif_names.append(out_dir + motif_name + "_pssm.txt")
break
for i in clusters4:
if first_cdr in clusters4[i]:
motif_name = str(c) + "_0.4"
generate_motif_from_cluster(motif_name, clusters4[i], out_dir)
motif_names.append(out_dir + motif_name + "_pssm.txt")
break
return([clust, clusters3[j], clusters4[i]])
else:
f = open(out_dir + "cluster_progression_results.txt", "w")
correct_n = True
# extract all clusters of correct size and process
for c in clusters2:
if n:
if len(clusters2[c]) == n:
correct_n = True
else:
correct_n = False
if correct_n and len(clusters2[c]) >= 2:
f.write("\n\noriginal cluster, at distance 0.2: ")
f.write(str(clusters2[c]))
f.write("\ncluster number " + str(c))
# process cluster - generate motif
motif_name = str(c) + "_0.2"
generate_motif_from_cluster(motif_name, clusters2[c], out_dir)
motif_names.append(out_dir + motif_name + "_pssm.txt")
# get first CDR - all unique, so can use to find how the cluster evolved
first_cdr = clusters2[c][0]
for j in clusters3:
if first_cdr in clusters3[j]:
f.write("\ncluster at distance 0.3: ")
f.write(str(clusters3[j]))
motif_name = str(c) + "_0.3"
generate_motif_from_cluster(motif_name, clusters3[j], out_dir)
motif_names.append(out_dir + motif_name + "_pssm.txt")
break
for i in clusters4:
if first_cdr in clusters4[i]:
f.write("\ncluster at distance 0.4: ")
f.write(str(clusters4[i]))
motif_name = str(c) + "_0.4"
generate_motif_from_cluster(motif_name, clusters4[i], out_dir)
motif_names.append(out_dir + motif_name + "_pssm.txt")
break
f.close()
return motif_names