comparison blastxmlToGff3.py @ 0:804a93e87cc8 draft

planemo upload for repository https://github.com/Yating-L/jbrowse_hub commit f22711ea7a464bdaf4d5aaea07f2eacf967aa66e-dirty
author yating-l
date Wed, 12 Apr 2017 17:41:55 -0400
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:804a93e87cc8
1 #!/usr/bin/env python
2
3
4 from Bio.Blast import NCBIXML
5 from collections import OrderedDict
6 import utils
7
8
9 def align2cigar(hsp_query, hsp_reference):
10 """
11 Build CIGAR representation from an hsp_query
12 input:
13 hsp_query
14 hsp_sbjct
15 output:
16 CIGAR string
17 """
18 query = hsp_query
19 ref = hsp_reference
20 # preType, curType:
21 # 'M' represents match,
22 # 'I' represents insert a gap into the reference sequence,
23 # 'D' represents insert a gap into the target (delete from reference)
24 # some ideas of this algin2cigar function are coming from
25 # https://gist.github.com/ozagordi/099bdb796507da8d9426
26 prevType = 'M'
27 curType = 'M'
28 count = 0
29 cigar = []
30 num = len(query)
31 for i in range(num):
32 if query[i] == '-':
33 curType = 'D'
34 elif ref[i] == '-':
35 curType = 'I'
36 else:
37 curType = 'M'
38 if curType == prevType:
39 count += 1
40 else:
41 cigar.append('%s%d' % (prevType, count))
42 prevType = curType
43 count = 1
44 cigar.append('%s%d' % (curType, count))
45 return ' '.join(cigar)
46
47 def gff3_writer(blast_records, gff3_file):
48 gff3 = open(gff3_file, 'a')
49 gff3.write("##gff-version 3\n")
50 seq_regions = dict()
51 for blast_record in blast_records:
52 query_name = blast_record.query.split(" ")[0]
53 source = blast_record.application
54 method = blast_record.matrix
55 for alignment in blast_record.alignments:
56 group = {
57 "parent_field" : OrderedDict(),
58 "parent_attribute" : OrderedDict(),
59 "alignments" : []
60 }
61 title = alignment.title.split(" ")
62 contig_name = title[len(title) - 1]
63 length = alignment.length
64 group['parent_field']['seqid'] = contig_name
65 group['parent_field']['source'] = source
66 group['parent_field']['type'] = 'match'
67 group['parent_attribute']['ID'] = contig_name + '_' + query_name
68 group['parent_attribute']['method'] = method
69 group['parent_attribute']['length'] = length
70 if contig_name not in seq_regions:
71 gff3.write("##sequence-region " + contig_name + ' 1 ' + str(length) + '\n')
72 seq_regions[contig_name] = length
73 match_num = 0
74 coords = [length, 0]
75 for hsp in alignment.hsps:
76 hsp_align = {}
77 field = OrderedDict()
78 attribute = OrderedDict()
79 ref = hsp.sbjct
80 query = hsp.query
81 field['seqid'] = contig_name
82 field['source'] = source
83 field['type'] = 'match_part'
84
85 field['start'] = hsp.sbjct_start
86 if field['start'] < coords[0]:
87 coords[0] = field['start']
88 ref_length = len(ref.replace('-', ''))
89 # if run tblastn, the actual length of reference should be multiplied by 3
90 if source.lower() == "tblastn":
91 ref_length *= 3
92 field['end'] = field['start'] + ref_length - 1
93 if field['end'] > coords[1]:
94 coords[1] = field['end']
95 field['score'] = hsp.score
96 #decide if the alignment in the same strand or reverse strand
97 #reading frame
98 # (+, +), (0, 0), (-, -) => +
99 # (+, -), (-, +) => -
100 if hsp.frame[1] * hsp.frame[0] > 0:
101 field['strand'] = '+'
102 elif hsp.frame[1] * hsp.frame[0] < 0:
103 field['strand'] = '-'
104 else:
105 if hsp.frame[0] + hsp.frame[1] >= 0:
106 field['strand'] = '+'
107 else:
108 field['strand'] = '-'
109 field['phase'] = '.'
110
111 target_start = hsp.query_start
112 target_len = len(query.replace('-', ''))
113 # if run blastx, the actual length of query should be multiplied by 3
114 if source.lower() == "blastx":
115 target_len *= 3
116 target_end = target_start + target_len -1
117 attribute['ID'] = group['parent_attribute']['ID'] + '_match_' + str(match_num)
118 attribute['Parent'] = group['parent_attribute']['ID']
119 attribute['Target'] = query_name + " " + str(target_start) + " " + str(target_end)
120 attribute['Gap'] = align2cigar(query, ref)
121 #store the query sequence and match string in the file in order to display alignment with BlastAlignment plugin
122 attribute['subject'] = hsp.sbjct
123 attribute['query'] = hsp.query
124 attribute['match'] = hsp.match
125 attribute['gaps'] = attribute['match'].count(' ')
126 similar = attribute['match'].count('+')
127 attribute['identities'] = len(attribute['match']) - similar - attribute['gaps']
128 attribute['positives'] = attribute['identities'] + similar
129 attribute['expect'] = hsp.expect
130 # show reading frame attribute only if the frame is not (0, 0)
131 attribute['frame'] = hsp.frame[1]
132 match_num += 1
133 hsp_align['field'] = field
134 hsp_align['attribute'] = attribute
135 group['alignments'].append(hsp_align)
136 group['parent_field']['start'] = coords[0]
137 group['parent_field']['end'] = coords[1]
138 group['parent_field']['score'] = group['parent_field']['strand'] = group['parent_field']['phase'] = '.'
139 group['parent_attribute']['match_num'] = match_num
140 group['alignments'].sort(key=lambda x: (x['field']['start'], x['field']['end']))
141 utils.write_features(group['parent_field'], group['parent_attribute'], gff3)
142 prev_end = -1
143 for align in group['alignments']:
144 overlap = ''
145 if align['field']['start'] <= prev_end:
146 overlap += str(align['field']['start']) + ',' + str(prev_end)
147 prev_end = align['field']['end']
148 align['attribute']['overlap'] = overlap
149 utils.write_features(align['field'], align['attribute'], gff3)
150 gff3.close()
151
152 def blastxml2gff3(xml_file, gff3_file):
153 result_handle = open(xml_file)
154 blast_records = NCBIXML.parse(result_handle)
155 gff3_writer(blast_records, gff3_file)
156
157 if __name__ == "__main__":
158 blastxml2gff3("../dbia3/raw/tblastn_dmel-hits-translation-r6.11.fa_vs_nucleotide_BLAST_database_from_data_3.blastxml", "gff3.txt")
159