comparison helper.py @ 5:6e589f267c14

Uploaded
author devteam
date Tue, 04 Nov 2014 12:15:19 -0500
parents
children
comparison
equal deleted inserted replaced
4:619e0fcd9126 5:6e589f267c14
1 #!/usr/bin/env python
2 """
3 Common utility functions
4 """
5
6 import os
7 import re
8 import sys
9 import gzip
10 import bz2
11 import numpy
12
13 def init_gene():
14 """
15 Initializing the gene structure
16 """
17
18 gene_det = [('id', 'f8'),
19 ('anno_id', numpy.dtype),
20 ('confgenes_id', numpy.dtype),
21 ('name', 'S25'),
22 ('source', 'S25'),
23 ('gene_info', numpy.dtype),
24 ('alias', 'S15'),
25 ('name2', numpy.dtype),
26 ('strand', 'S2'),
27 ('score', 'S15'),
28 ('chr', 'S15'),
29 ('chr_num', numpy.dtype),
30 ('paralogs', numpy.dtype),
31 ('start', 'f8'),
32 ('stop', 'f8'),
33 ('transcripts', numpy.dtype),
34 ('transcript_type', numpy.dtype),
35 ('transcript_info', numpy.dtype),
36 ('transcript_status', numpy.dtype),
37 ('transcript_valid', numpy.dtype),
38 ('exons', numpy.dtype),
39 ('exons_confirmed', numpy.dtype),
40 ('cds_exons', numpy.dtype),
41 ('utr5_exons', numpy.dtype),
42 ('utr3_exons', numpy.dtype),
43 ('tis', numpy.dtype),
44 ('tis_conf', numpy.dtype),
45 ('tis_info', numpy.dtype),
46 ('cdsStop', numpy.dtype),
47 ('cdsStop_conf', numpy.dtype),
48 ('cdsStop_info', numpy.dtype),
49 ('tss', numpy.dtype),
50 ('tss_info', numpy.dtype),
51 ('tss_conf', numpy.dtype),
52 ('cleave', numpy.dtype),
53 ('cleave_info', numpy.dtype),
54 ('cleave_conf', numpy.dtype),
55 ('polya', numpy.dtype),
56 ('polya_info', numpy.dtype),
57 ('polya_conf', numpy.dtype),
58 ('is_alt', 'f8'),
59 ('is_alt_spliced', 'f8'),
60 ('is_valid', numpy.dtype),
61 ('transcript_complete', numpy.dtype),
62 ('is_complete', numpy.dtype),
63 ('is_correctly_gff3_referenced', 'S5'),
64 ('splicegraph', numpy.dtype) ]
65
66 return gene_det
67
68 def open_file(fname):
69 """
70 Open the file (supports .gz .bz2) and returns the handler
71
72 @args fname: input file name for reading
73 @type fname: str
74 """
75
76 try:
77 if os.path.splitext(fname)[1] == ".gz":
78 FH = gzip.open(fname, 'rb')
79 elif os.path.splitext(fname)[1] == ".bz2":
80 FH = bz2.BZ2File(fname, 'rb')
81 else:
82 FH = open(fname, 'rU')
83 except Exception as error:
84 sys.exit(error)
85
86 return FH
87
88 def add_CDS_phase(strand, cds):
89 """
90 Calculate CDS phase and add to the CDS exons
91
92 @args strand: feature strand information
93 @type strand: +/-
94 @args cds: coding exon coordinates
95 @type cds: numpy array [[int, int, int]]
96 """
97
98 cds_region, cds_flag = [], 0
99 if strand == '+':
100 for cdspos in cds:
101 if cds_flag == 0:
102 cdspos = (cdspos[0], cdspos[1], 0)
103 diff = (cdspos[1]-(cdspos[0]-1))%3
104 else:
105 xy = 0
106 if diff == 0:
107 cdspos = (cdspos[0], cdspos[1], 0)
108 elif diff == 1:
109 cdspos = (cdspos[0], cdspos[1], 2)
110 xy = 2
111 elif diff == 2:
112 cdspos = (cdspos[0], cdspos[1], 1)
113 xy = 1
114 diff = ((cdspos[1]-(cdspos[0]-1))-xy)%3
115 cds_region.append(cdspos)
116 cds_flag = 1
117 elif strand == '-':
118 cds.reverse()
119 for cdspos in cds:
120 if cds_flag == 0:
121 cdspos = (cdspos[0], cdspos[1], 0)
122 diff = (cdspos[1]-(cdspos[0]-1))%3
123 else:
124 xy = 0
125 if diff == 0:
126 cdspos = (cdspos[0], cdspos[1], 0)
127 elif diff == 1:
128 cdspos = (cdspos[0], cdspos[1], 2)
129 xy = 2
130 elif diff == 2:
131 cdspos = (cdspos[0], cdspos[1], 1)
132 xy = 1
133 diff = ((cdspos[1]-(cdspos[0]-1))-xy)%3
134 cds_region.append(cdspos)
135 cds_flag = 1
136 cds_region.reverse()
137 return cds_region
138
139 def buildUTR(cc, ec, strand):
140 """
141 Build UTR regions from a given set of CDS and exon coordiantes of a gene
142
143 @args cc: coding exon coordinates
144 @type cc: numpy array [[int, int, int]]
145 @args ec: exon coordinates
146 @type ec: numpy array [[int, int]]
147 @args strand: feature strand information
148 @type strand: +/-
149 """
150
151 utr5 = []
152 utr3 = []
153 if strand == '+':
154 cds_s = cc[0][0]
155 for ex in ec:
156 if ex[0] <= cds_s and cds_s <= ex[1]:
157 if ex[0] != cds_s:utr5.append((ex[0], cds_s-1))
158 break
159 else:
160 utr5.append(ex)
161 cds_e = cc[-1][1]
162 for i in range(len(ec)):
163 i += 1
164 if ec[-i][0] <= cds_e and cds_e <= ec[-i][1]:
165 if ec[-i][1] != cds_e:utr3.append((cds_e +1, ec[-i][1]))
166 break
167 else:
168 utr3.append(ec[-i])
169 utr3.reverse()
170 elif strand == '-':
171 cds_s = cc[-1][1]
172 for i in range(len(ec)):
173 i += 1
174 if ec[-i][0] <= cds_s and cds_s <= ec[-i][1]:
175 if ec[-i][1] != cds_s:utr5.append((cds_s+1, ec[-i][1]))
176 break
177 else:
178 utr5.append(ec[-i])
179 utr5.reverse()
180 cds_e = cc[0][0]
181 for ex in ec:
182 if ex[0] <= cds_e and cds_e <= ex[1]:
183 if ex[0] != cds_e:utr3.append((ex[0], cds_e-1))
184 break
185 else:
186 utr3.append(ex)
187 return utr5, utr3
188
189 def make_Exon_cod(strand_p, five_p_utr, cds_cod, three_p_utr):
190 """
191 Create exon cordinates from UTR's and CDS region
192
193 @args strand_p: feature strand information
194 @type strand_p: +/-
195 @args five_p_utr: five prime utr exon coordinates
196 @type five_p_utr: numpy array [[int, int]]
197 @args cds_cod: coding exon coordinates
198 @type cds_cod: numpy array [[int, int, int]]
199 @args three_p_utr: three prime utr exon coordinates
200 @type three_p_utr: numpy array [[int, int]]
201 """
202
203 exon_pos = []
204 if strand_p == '+':
205 utr5_start, utr5_end = 0, 0
206 if five_p_utr != []:
207 utr5_start, utr5_end = five_p_utr[-1][0], five_p_utr[-1][1]
208 cds_5start, cds_5end = cds_cod[0][0], cds_cod[0][1]
209 jun_exon = []
210 if cds_5start-utr5_end == 0 or cds_5start-utr5_end == 1:
211 jun_exon = [utr5_start, cds_5end]
212 if len(cds_cod) == 1:
213 five_prime_flag = 0
214 if jun_exon != []:
215 five_p_utr = five_p_utr[:-1]
216 five_prime_flag = 1
217 for utr5 in five_p_utr:
218 exon_pos.append(utr5)
219 jun_exon = []
220 utr3_start, utr3_end = 0, 0
221 if three_p_utr != []:
222 utr3_start = three_p_utr[0][0]
223 utr3_end = three_p_utr[0][1]
224 if utr3_start-cds_5end == 0 or utr3_start-cds_5end == 1:
225 jun_exon = [cds_5start, utr3_end]
226 three_prime_flag = 0
227 if jun_exon != []:
228 cds_cod = cds_cod[:-1]
229 three_p_utr = three_p_utr[1:]
230 three_prime_flag = 1
231 if five_prime_flag == 1 and three_prime_flag == 1:
232 exon_pos.append([utr5_start, utr3_end])
233 if five_prime_flag == 1 and three_prime_flag == 0:
234 exon_pos.append([utr5_start, cds_5end])
235 cds_cod = cds_cod[:-1]
236 if five_prime_flag == 0 and three_prime_flag == 1:
237 exon_pos.append([cds_5start, utr3_end])
238 for cds in cds_cod:
239 exon_pos.append(cds)
240 for utr3 in three_p_utr:
241 exon_pos.append(utr3)
242 else:
243 if jun_exon != []:
244 five_p_utr = five_p_utr[:-1]
245 cds_cod = cds_cod[1:]
246 for utr5 in five_p_utr:
247 exon_pos.append(utr5)
248 exon_pos.append(jun_exon) if jun_exon != [] else ''
249 jun_exon = []
250 utr3_start, utr3_end = 0, 0
251 if three_p_utr != []:
252 utr3_start = three_p_utr[0][0]
253 utr3_end = three_p_utr[0][1]
254 cds_3start = cds_cod[-1][0]
255 cds_3end = cds_cod[-1][1]
256 if utr3_start-cds_3end == 0 or utr3_start-cds_3end == 1:
257 jun_exon = [cds_3start, utr3_end]
258 if jun_exon != []:
259 cds_cod = cds_cod[:-1]
260 three_p_utr = three_p_utr[1:]
261 for cds in cds_cod:
262 exon_pos.append(cds)
263 exon_pos.append(jun_exon) if jun_exon != [] else ''
264 for utr3 in three_p_utr:
265 exon_pos.append(utr3)
266 elif strand_p == '-':
267 utr3_start, utr3_end = 0, 0
268 if three_p_utr != []:
269 utr3_start = three_p_utr[-1][0]
270 utr3_end = three_p_utr[-1][1]
271 cds_3start = cds_cod[0][0]
272 cds_3end = cds_cod[0][1]
273 jun_exon = []
274 if cds_3start-utr3_end == 0 or cds_3start-utr3_end == 1:
275 jun_exon = [utr3_start, cds_3end]
276 if len(cds_cod) == 1:
277 three_prime_flag = 0
278 if jun_exon != []:
279 three_p_utr = three_p_utr[:-1]
280 three_prime_flag = 1
281 for utr3 in three_p_utr:
282 exon_pos.append(utr3)
283 jun_exon = []
284 (utr5_start, utr5_end) = (0, 0)
285 if five_p_utr != []:
286 utr5_start = five_p_utr[0][0]
287 utr5_end = five_p_utr[0][1]
288 if utr5_start-cds_3end == 0 or utr5_start-cds_3end == 1:
289 jun_exon = [cds_3start, utr5_end]
290 five_prime_flag = 0
291 if jun_exon != []:
292 cds_cod = cds_cod[:-1]
293 five_p_utr = five_p_utr[1:]
294 five_prime_flag = 1
295 if three_prime_flag == 1 and five_prime_flag == 1:
296 exon_pos.append([utr3_start, utr5_end])
297 if three_prime_flag == 1 and five_prime_flag == 0:
298 exon_pos.append([utr3_start, cds_3end])
299 cds_cod = cds_cod[:-1]
300 if three_prime_flag == 0 and five_prime_flag == 1:
301 exon_pos.append([cds_3start, utr5_end])
302 for cds in cds_cod:
303 exon_pos.append(cds)
304 for utr5 in five_p_utr:
305 exon_pos.append(utr5)
306 else:
307 if jun_exon != []:
308 three_p_utr = three_p_utr[:-1]
309 cds_cod = cds_cod[1:]
310 for utr3 in three_p_utr:
311 exon_pos.append(utr3)
312 if jun_exon != []:
313 exon_pos.append(jun_exon)
314 jun_exon = []
315 (utr5_start, utr5_end) = (0, 0)
316 if five_p_utr != []:
317 utr5_start = five_p_utr[0][0]
318 utr5_end = five_p_utr[0][1]
319 cds_5start = cds_cod[-1][0]
320 cds_5end = cds_cod[-1][1]
321 if utr5_start-cds_5end == 0 or utr5_start-cds_5end == 1:
322 jun_exon = [cds_5start, utr5_end]
323 if jun_exon != []:
324 cds_cod = cds_cod[:-1]
325 five_p_utr = five_p_utr[1:]
326 for cds in cds_cod:
327 exon_pos.append(cds)
328 if jun_exon != []:
329 exon_pos.append(jun_exon)
330 for utr5 in five_p_utr:
331 exon_pos.append(utr5)
332 return exon_pos