design.py 14.8 KB
Newer Older
Bollen's avatar
init  
Bollen committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
from builtins import (ascii, bytes, chr, dict, filter, hex, input,
                      map, next, oct, open, pow, range, round,
                      str, super, zip)
__author__ = 'ahbbollen'

from tempfile import TemporaryFile, NamedTemporaryFile
from subprocess import check_call
import os
import re
import warnings
from time import sleep

from Bio import SeqIO
from Bio.Seq import Seq
from Bio.SeqRecord import SeqRecord
from Bio.Alphabet import generic_dna

from fastools.fastools import get_reference
from pyfaidx import Fasta
Sander Bollen's avatar
Sander Bollen committed
20
from pysam import AlignmentFile
Bollen's avatar
init  
Bollen committed
21
22

from .models import Primer, Region, BlatLine
Sander Bollen's avatar
Sander Bollen committed
23
from .utils import NoPrimersException, calc_gc, NEW_VCF, generate_fastq_from_primers
Bollen's avatar
init  
Bollen committed
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

PRIMER3_SCRIPT = os.path.join(os.path.join(os.path.dirname(__file__) ,"static"), 'getprimers.sh')


def get_sequence_fasta(region, reference=None, padding=True):
    ref = Fasta(reference)
    if "chr" not in list(ref.keys())[0] and "chr" in region.chr:
        chrom = region.chr.split("chr")[1]
    elif "chr" not in region.chr and "chr" in list(ref.keys())[0]:
        chrom = "chr" + region.chr
    else:
        chrom = region.chr

    if not padding:
        return ref[chrom][region.start:region.stop].seq
    else:
        return ref[chrom][region.start_w_padding:region.stop_w_padding].seq


def run_primer3(sequence, region, padding=True,
                primer3_script=PRIMER3_SCRIPT, product_size="200-450",
                n_primers=4, prim3_exe=None):
    if padding:
        target_start = region.padding_left
        target_len = len(sequence) - region.padding_left - region.padding_right
    else:
        target_start = 1
        target_len = len(sequence)

    target = ",".join(map(str, [target_start, target_len]))
    opt_size = str(sum(map(int, product_size.split("-"))) / 2)

    args = [primer3_script, product_size, target, sequence, target, opt_size, prim3_exe]
    retval = check_call(args)
    if retval != 0:
        raise ValueError("Primer3 crashed")

    # now read example.for and example.rev

    with open("example.for", "rb") as forward, open("example.rev", "rb") as reverse:
        forwards = _sanitize_p3(forward)
        reverses = _sanitize_p3(reverse)

    forwards, reverses = _get_shortest(forwards, reverses, n_primers)

    if len(forwards) == 0 or len(reverses) == 0:
        raise NoPrimersException("No acceptable primers could be found. Try increasing the padding")

    primers = [Primer.from_p3(x, y, sequence, region.chr, region.start) for x, y in zip(forwards, reverses)]
    return primers


Sander Bollen's avatar
Sander Bollen committed
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
def aln_primers(primers, bwa_exe=None, samtools_exe=None, ref=None, output_bam=None):
    """
    Align primers with BWA. 
    This only works with BWA-ALN due to the short read length. 
    :param primers: List of primers
    :param bwa_exe: Path to BWA
    :param samtools_exe: Path to samtools
    :param ref: Path to reference fasta
    :param output_bam: Path to final bam file
    :return: instance of pysam.AlignmentFile 
    """
    fq1 = NamedTemporaryFile()
    fq2 = NamedTemporaryFile()
    sai1 = NamedTemporaryFile()
    sai2 = NamedTemporaryFile()

    generate_fastq_from_primers(primers, fq1.name, fq2.name)
    aln_args1 = [bwa_exe, 'aln', ref, fq1.name]
    aln_args2 = [bwa_exe, 'aln', ref, fq2.name]

    r = check_call(aln_args1, stdout=sai1)
    if r != 0:
        raise ValueError("bwa aln crashed with error code {0}".format(r))
    r = check_call(aln_args2, stdout=sai2)
    if r != 0:
        raise ValueError("bwa aln crashed with error code {0}".format(r))

    sam = NamedTemporaryFile()
    sam_args = [bwa_exe, 'sampe', ref, sai1.name, sai2.name, fq1.name, fq2.name]
    r = check_call(sam_args, stdout=sam)
    if r != 0:
        raise ValueError("bwa sampe crashed with error code {0}".format(r))

    bam = NamedTemporaryFile()
    bam_args = [samtools_exe, 'view', '-Shb', sam.name]
    r = check_call(bam_args, stdout=bam)
    if r != 0:
        raise ValueError("samtools view crashed with error code {0}".format(r))

    final_args = [samtools_exe, 'sort', "-f", bam.name, output_bam]
    r = check_call(final_args)
    if r != 0:
        raise ValueError("samtools sort crashed with error code {0}".format(r))

    index_args = [samtools_exe, "index", output_bam]
    r = check_call(index_args)
    if r != 0:
        raise ValueError("samtools index crashed with error code {0}".format(r))

    # cleanup
    for i in [fq1, fq2, sai1, sai2, sam, bam]:
        i.close()

    return AlignmentFile(output_bam, "rb")


def _get_match_fraction(aligned_segment):
    """Get the fraction of a read matching the reference"""
    matching_bases = aligned_segment.cigartuples[0][1]
    return float(matching_bases)/aligned_segment.query_length


def _read_on_same_chrom(aligned_segment, variant=None, region=None):
    """Return true if read is on same chromosome as variant or region"""
    read_chrom = aligned_segment.reference_name.split("chr")[-1]
    if variant is not None:
        variant_chrom = variant.chromosome.split("chr")[-1]
    elif region is not None:
        variant_chrom = region.chr.split("chr")[-1]
    else:
        raise ValueError
    return read_chrom == variant_chrom
Bollen's avatar
init  
Bollen committed
148
149


Sander Bollen's avatar
Sander Bollen committed
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
def _has_alternative_alignments(aligned_segment):
    """Return true if read has alternative alignments"""
    tags = aligned_segment.get_tags()
    return "XA" in [x[0] for x in tags]


def create_primer_from_pair(read1, read2, position=0):
    """Create primer from read pair"""
    return Primer(
        chromosome=read1.reference_name,
        position=position,
        left=read1.query_sequence,
        right=read2.query_sequence,
        left_pos=read1.reference_start,
        right_pos=read2.reference_start,
        left_len=read1.query_length,
        right_len=read2.query_length,
        left_gc=calc_gc(read1.query_sequence),
        right_gc=calc_gc(read2.query_sequence)
    )


def create_primers_bwa(bam_handle, variant=None, region=None):
    """
    Find correct pairs in a bam file containing primers for a variant or region    
    This requires to hold the bam file in memory. Use only with small bam files
    :param bam_handle: pysam.AlignmentFile instance
    :param variant: variant
    :param region: region 
    :return: generator of primers
    """
    if variant is not None and region is not None:
        raise ValueError("Either variant _or_ region must be set")
    elif variant is None and region is None:
        raise ValueError("Either variant _or_ region must be set")

    pairs = {}  # bucket to hold pairs
    for read in bam_handle:
        if read.is_read1:
            if read.query_name in pairs:
                pairs[read.query_name].update({"r1": read})
            else:
                pairs[read.query_name] = {"r1": read}
        elif read.is_read2:
            if read.query_name in pairs:
                pairs[read.query_name].update({"r2": read})
            else:
                pairs[read.query_name] = {"r2": read}

    for pair in pairs.values():
        # must match at least 90%
        # must be on same chromosome as variant
        # may not have alternative alignments
        # pair must be complete
        if "r1" not in pair or "r2" not in pair:
            continue
        if _get_match_fraction(pair['r1']) < 0.9:
            continue
        if _get_match_fraction(pair['r2']) < 0.9:
            continue

        if variant is not None:
            if not _read_on_same_chrom(pair['r1'], variant=variant):
                continue
            if not _read_on_same_chrom(pair['r2'], variant=variant):
                continue
Bollen's avatar
init  
Bollen committed
216
        else:
Sander Bollen's avatar
Sander Bollen committed
217
218
219
220
            if not _read_on_same_chrom(pair['r1'], region=region):
                continue
            if not _read_on_same_chrom(pair['r2'], region=region):
                continue
Bollen's avatar
init  
Bollen committed
221

Sander Bollen's avatar
Sander Bollen committed
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
        if _has_alternative_alignments(pair['r1']):
            continue
        if _has_alternative_alignments(pair['r2']):
            continue

        if variant is not None:
            yield create_primer_from_pair(pair['r1'], pair['r2'], variant.position_g_start)
        else:
            position = int(region.start) + int((float((int(region.stop) - int(region.start)))/2))
            yield create_primer_from_pair(pair['r1'], pair['r2'], position)


def find_best_bwa(bam, variant=None, region=None, accept_snp=False, field=None, max_freq=None, dbsnp=None):
    primers = []
    for primer in create_primers_bwa(bam, variant=variant, region=region):
        if not accept_snp and max_freq is not None:
            prim = find_snps(primer, dbsnp, field)
            if prim.snp_freq <= max_freq:
                primers.append(prim)
        else:
            primers.append(primer)
Bollen's avatar
init  
Bollen committed
243
244
    if len(primers) == 0:
        raise NoPrimersException("No suitable primers could be detected")
Sander Bollen's avatar
Sander Bollen committed
245
246
    else:
        return primers
Bollen's avatar
init  
Bollen committed
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
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353


def _freq_in_query(query, field):
    """
    Get amount of records and frequencies for a query (= reader iterator)
    :param query: VCF reader iterator
    :param field: the field to fetch
    :return:
    """
    n = 0
    freqs = []
    for record in query:
        n += 1
        try:
            freqs += list(map(float, record.INFO[field]))
        except ValueError:
            freqs += [0]
        except KeyError:
            freqs += [0]
    return n, freqs


def find_snps(primer, db_snp=None, field="AF"):
    try:
        import vcf
    except:
        return None

    reader = vcf.Reader(filename=db_snp)

    left_end = int(primer.left_pos) + len(primer.left)
    right_end = int(primer.right_pos) + len(primer.right)

    contigs = list(reader.contigs.keys())
    if "chr" in contigs[0] and not "chr" in primer.chromosome:
        chrom = "chr" + primer.chromosome
    elif "chr" in primer.chromosome and not "chr" in contigs[0]:
        chrom = primer.chromosome.split("chr")[1]
    else:
        chrom = primer.chromosome

    if NEW_VCF:
        query = reader.fetch(chrom, int(primer.left_pos), left_end)
    else:
        query = reader.fetch(chrom, int(primer.left_pos) + 1, left_end)

    left_n, left_freqs = _freq_in_query(query, field)

    # can't have two pysam queries open at the same time, so have to do this this way
    if NEW_VCF:
        right_query = reader.fetch(chrom, int(primer.right_pos), right_end)
    else:
        right_query = reader.fetch(chrom, int(primer.right_pos) + 1, right_end)

    right_n, right_freqs = _freq_in_query(right_query, field)
    n = left_n + right_n
    freqs = right_freqs + left_freqs

    if n > 0:
        primer.contains_SNP = True
        primer.contains_freq_SNP = True
        if len(freqs) > 0:
            primer.snp_freq = max(freqs)
    return primer


def _sanitize_p3(handle):
    # sanitize p3 output
    # first line should always be removed:
    _ = next(handle)
    return [x for x in handle if "#" not in x.decode()]


def _get_shortest(x, y, max_v):
    small = min(len(x), len(y), max_v)
    return x[:small], y[:small]


def chop_region(region, size):
    """
    Chop a region in multiple regions with max size `size`
    Child regions inherit the padding of their parents
    :param region: the region to be chopped
    :param size: integer
    :return: List[Region]
    """
    if len(region) <= size:
        return [region]
    first = Region(chromosome=region.chr, start=region.start, stop=region.start+size,
                   acc_nr=region.acc_nr, padding_left=region.padding_left, padding_right=region.padding_right,
                   other=region.other_information)
    regions = [first]
    while sum([len(x) for x in regions]) < len(region):
        last = regions[-1]
        if last.stop + size >= region.stop:
            stop = region.stop
        else:
            stop = last.stop + size

        nex = Region(chromosome=region.chr, start=last.stop, stop=stop, acc_nr=region.acc_nr,
                     padding_left=region.padding_left, padding_right=region.padding_right,
                     other=region.other_information)
        regions.append(nex)
    return regions


def get_primer_from_region(region, reference, product_size, n_prims,
Sander Bollen's avatar
Sander Bollen committed
354
355
                           bwa_exe, samtools_exe, primer3_exe,
                           output_bam=None, dbsnp=None, field=None,
Bollen's avatar
init  
Bollen committed
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
                           max_freq=None, strict=False, min_margin=10):

    min_length, max_length = list(map(int, product_size.split("-")))
    regions = chop_region(region, min_length)
    if any([x.size(True) > max_length for x in regions]):
        warnings.warn("Current padding results in larger regions than allowed")

    primers = []
    return_regions = []

    for reg in regions:
        sequence = get_sequence_fasta(reg, reference=reference)
        raw_primers = run_primer3(sequence, reg, padding=True,
                                  product_size=product_size,
                                  n_primers=n_prims, prim3_exe=primer3_exe)
Sander Bollen's avatar
Sander Bollen committed
371
372
373
374
375
        bam = aln_primers(raw_primers, bwa_exe=bwa_exe, samtools_exe=samtools_exe,
                          ref=reference, output_bam=output_bam)
        prims = find_best_bwa(bam, region=reg, dbsnp=dbsnp, field=field, max_freq=max_freq)
        tmp_reg = []
        tmp_prim = []
Bollen's avatar
init  
Bollen committed
376
377

        # filter out primers too close to the variant
Sander Bollen's avatar
Sander Bollen committed
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
        for primer in prims:
            if not (int(primer.left_pos) + int(primer.left_len) + min_margin) < int(reg.start):
                continue
            if not (int(primer.right_pos) - min_margin) > int(reg.stop):
                continue

            n_region = Region(start=int(primer.left_pos), stop=int(primer.right_pos)+len(primer.right),
                              chromosome=primer.chromosome, padding_left=0, padding_right=0,
                              acc_nr="NA", other="NA")
            primer.fragment_sequence = get_sequence_fasta(n_region, reference=reference, padding=False)
            if strict and len(primer.fragment_sequence) > max_length:
                continue
            else:
                tmp_prim.append(primer)
                tmp_reg.append(reg)
        if len(tmp_prim) > 0 and len(tmp_prim) > 0:
            primers.append(tmp_prim[0])
Bollen's avatar
init  
Bollen committed
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
            return_regions.append(reg)
    if len(primers) == 0 or len(return_regions) == 0:
        raise NoPrimersException
    return return_regions, primers


def create_left_prim(primer, reference):
    primer_r_pos = primer.position + int(primer.right_pos)
    n_region = Region(start=primer_r_pos - len(primer.right), stop=primer_r_pos,
                      chromosome=primer.chromosome, padding_left=0, padding_right=0,
                      acc_nr="NA", other="NA")
    next_left_seq = get_sequence_fasta(n_region, reference=reference, padding=False)
    next_left_pos = primer_r_pos - len(primer.right)
    next_left_gc = calc_gc(next_left_seq)
    next_left = Primer()
    next_left.left = next_left_seq
    next_left.left_gc = next_left_gc
    next_left.left_pos = next_left_pos
    next_left.left_len = len(next_left_seq)
    next_left.left_name = '.'.join([primer.chromosome, str(primer.position)]) + "_left"
    next_left.chromosome = primer.chromosome
    next_left.position = primer_r_pos

    return next_left