Mercurial > repos > iuc > vsnp_build_tables

comparison vsnp_add_zero_coverage.py @ 1:0bc0009f9ea0 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/vsnp commit 6a0c9a857c1f4638ef18e106b1f8c0681303acc5"
author iuc
date Sun, 27 Sep 2020 10:08:14 +0000
parents
children a52b819aa990
comparison
equal deleted inserted replaced
0:5e258fba246c 1:0bc0009f9ea0
1 #!/usr/bin/env python
2
3 import argparse
4 import multiprocessing
5 import os
6 import pandas
7 import pysam
8 import queue
9 import re
10 import shutil
11 from Bio import SeqIO
12
13 INPUT_BAM_DIR = 'input_bam_dir'
14 INPUT_VCF_DIR = 'input_vcf_dir'
15 OUTPUT_VCF_DIR = 'output_vcf_dir'
16 OUTPUT_METRICS_DIR = 'output_metrics_dir'
17
18
19 def get_base_file_name(file_path):
20 base_file_name = os.path.basename(file_path)
21 if base_file_name.find(".") > 0:
22 # Eliminate the extension.
23 return os.path.splitext(base_file_name)[0]
24 elif base_file_name.endswith("_vcf"):
25 # The "." character has likely
26 # changed to an "_" character.
27 return base_file_name.rstrip("_vcf")
28 return base_file_name
29
30
31 def get_coverage_and_snp_count(task_queue, reference, output_metrics, output_vcf, timeout):
32 while True:
33 try:
34 tup = task_queue.get(block=True, timeout=timeout)
35 except queue.Empty:
36 break
37 bam_file, vcf_file = tup
38 # Create a coverage dictionary.
39 coverage_dict = {}
40 coverage_list = pysam.depth(bam_file, split_lines=True)
41 for line in coverage_list:
42 chrom, position, depth = line.split('\t')
43 coverage_dict["%s-%s" % (chrom, position)] = depth
44 # Convert it to a data frame.
45 coverage_df = pandas.DataFrame.from_dict(coverage_dict, orient='index', columns=["depth"])
46 # Create a zero coverage dictionary.
47 zero_dict = {}
48 for record in SeqIO.parse(reference, "fasta"):
49 chrom = record.id
50 total_len = len(record.seq)
51 for pos in list(range(1, total_len + 1)):
52 zero_dict["%s-%s" % (str(chrom), str(pos))] = 0
53 # Convert it to a data frame with depth_x
54 # and depth_y columns - index is NaN.
55 zero_df = pandas.DataFrame.from_dict(zero_dict, orient='index', columns=["depth"])
56 coverage_df = zero_df.merge(coverage_df, left_index=True, right_index=True, how='outer')
57 # depth_x "0" column no longer needed.
58 coverage_df = coverage_df.drop(columns=['depth_x'])
59 coverage_df = coverage_df.rename(columns={'depth_y': 'depth'})
60 # Covert the NaN to 0 coverage and get some metrics.
61 coverage_df = coverage_df.fillna(0)
62 coverage_df['depth'] = coverage_df['depth'].apply(int)
63 total_length = len(coverage_df)
64 average_coverage = coverage_df['depth'].mean()
65 zero_df = coverage_df[coverage_df['depth'] == 0]
66 total_zero_coverage = len(zero_df)
67 total_coverage = total_length - total_zero_coverage
68 genome_coverage = "{:.2%}".format(total_coverage / total_length)
69 # Process the associated VCF input.
70 column_names = ["CHROM", "POS", "ID", "REF", "ALT", "QUAL", "FILTER", "INFO", "FORMAT", "Sample"]
71 vcf_df = pandas.read_csv(vcf_file, sep='\t', header=None, names=column_names, comment='#')
72 good_snp_count = len(vcf_df[(vcf_df['ALT'].str.len() == 1) & (vcf_df['REF'].str.len() == 1) & (vcf_df['QUAL'] > 150)])
73 base_file_name = get_base_file_name(vcf_file)
74 if total_zero_coverage > 0:
75 header_file = "%s_header.csv" % base_file_name
76 with open(header_file, 'w') as outfile:
77 with open(vcf_file) as infile:
78 for line in infile:
79 if re.search('^#', line):
80 outfile.write("%s" % line)
81 vcf_df_snp = vcf_df[vcf_df['REF'].str.len() == 1]
82 vcf_df_snp = vcf_df_snp[vcf_df_snp['ALT'].str.len() == 1]
83 vcf_df_snp['ABS_VALUE'] = vcf_df_snp['CHROM'].map(str) + "-" + vcf_df_snp['POS'].map(str)
84 vcf_df_snp = vcf_df_snp.set_index('ABS_VALUE')
85 cat_df = pandas.concat([vcf_df_snp, zero_df], axis=1, sort=False)
86 cat_df = cat_df.drop(columns=['CHROM', 'POS', 'depth'])
87 cat_df[['ID', 'ALT', 'QUAL', 'FILTER', 'INFO']] = cat_df[['ID', 'ALT', 'QUAL', 'FILTER', 'INFO']].fillna('.')
88 cat_df['REF'] = cat_df['REF'].fillna('N')
89 cat_df['FORMAT'] = cat_df['FORMAT'].fillna('GT')
90 cat_df['Sample'] = cat_df['Sample'].fillna('./.')
91 cat_df['temp'] = cat_df.index.str.rsplit('-', n=1)
92 cat_df[['CHROM', 'POS']] = pandas.DataFrame(cat_df.temp.values.tolist(), index=cat_df.index)
93 cat_df = cat_df[['CHROM', 'POS', 'ID', 'REF', 'ALT', 'QUAL', 'FILTER', 'INFO', 'FORMAT', 'Sample']]
94 cat_df['POS'] = cat_df['POS'].astype(int)
95 cat_df = cat_df.sort_values(['CHROM', 'POS'])
96 body_file = "%s_body.csv" % base_file_name
97 cat_df.to_csv(body_file, sep='\t', header=False, index=False)
98 if output_vcf is None:
99 output_vcf_file = os.path.join(OUTPUT_VCF_DIR, "%s.vcf" % base_file_name)
100 else:
101 output_vcf_file = output_vcf
102 with open(output_vcf_file, "w") as outfile:
103 for cf in [header_file, body_file]:
104 with open(cf, "r") as infile:
105 for line in infile:
106 outfile.write("%s" % line)
107 else:
108 if output_vcf is None:
109 output_vcf_file = os.path.join(OUTPUT_VCF_DIR, "%s.vcf" % base_file_name)
110 else:
111 output_vcf_file = output_vcf
112 shutil.copyfile(vcf_file, output_vcf_file)
113 bam_metrics = [base_file_name, "", "%4f" % average_coverage, genome_coverage]
114 vcf_metrics = [base_file_name, str(good_snp_count), "", ""]
115 if output_metrics is None:
116 output_metrics_file = os.path.join(OUTPUT_METRICS_DIR, "%s.tabular" % base_file_name)
117 else:
118 output_metrics_file = output_metrics
119 metrics_columns = ["File", "Number of Good SNPs", "Average Coverage", "Genome Coverage"]
120 with open(output_metrics_file, "w") as fh:
121 fh.write("# %s\n" % "\t".join(metrics_columns))
122 fh.write("%s\n" % "\t".join(bam_metrics))
123 fh.write("%s\n" % "\t".join(vcf_metrics))
124 task_queue.task_done()
125
126
127 def set_num_cpus(num_files, processes):
128 num_cpus = int(multiprocessing.cpu_count())
129 if num_files < num_cpus and num_files < processes:
130 return num_files
131 if num_cpus < processes:
132 half_cpus = int(num_cpus / 2)
133 if num_files < half_cpus:
134 return num_files
135 return half_cpus
136 return processes
137
138
139 if __name__ == '__main__':
140 parser = argparse.ArgumentParser()
141
142 parser.add_argument('--output_metrics', action='store', dest='output_metrics', required=False, default=None, help='Output metrics text file')
143 parser.add_argument('--output_vcf', action='store', dest='output_vcf', required=False, default=None, help='Output VCF file')
144 parser.add_argument('--reference', action='store', dest='reference', help='Reference dataset')
145 parser.add_argument('--processes', action='store', dest='processes', type=int, help='User-selected number of processes to use for job splitting')
146
147 args = parser.parse_args()
148
149 # The assumption here is that the list of files
150 # in both INPUT_BAM_DIR and INPUT_VCF_DIR are
151 # equal in number and named such that they are
152 # properly matched if the directories contain
153 # more than 1 file (i.e., hopefully the bam file
154 # names and vcf file names will be something like
155 # Mbovis-01D6_* so they can be # sorted and properly
156 # associated with each other).
157 bam_files = []
158 for file_name in sorted(os.listdir(INPUT_BAM_DIR)):
159 file_path = os.path.abspath(os.path.join(INPUT_BAM_DIR, file_name))
160 bam_files.append(file_path)
161 vcf_files = []
162 for file_name in sorted(os.listdir(INPUT_VCF_DIR)):
163 file_path = os.path.abspath(os.path.join(INPUT_VCF_DIR, file_name))
164 vcf_files.append(file_path)
165
166 multiprocessing.set_start_method('spawn')
167 queue1 = multiprocessing.JoinableQueue()
168 num_files = len(bam_files)
169 cpus = set_num_cpus(num_files, args.processes)
170 # Set a timeout for get()s in the queue.
171 timeout = 0.05
172
173 # Add each associated bam and vcf file pair to the queue.
174 for i, bam_file in enumerate(bam_files):
175 vcf_file = vcf_files[i]
176 queue1.put((bam_file, vcf_file))
177
178 # Complete the get_coverage_and_snp_count task.
179 processes = [multiprocessing.Process(target=get_coverage_and_snp_count, args=(queue1, args.reference, args.output_metrics, args.output_vcf, timeout, )) for _ in range(cpus)]
180 for p in processes:
181 p.start()
182 for p in processes:
183 p.join()
184 queue1.join()
185
186 if queue1.empty():
187 queue1.close()
188 queue1.join_thread()