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
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
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
|
#!/usr/bin/python3
# -*- coding: utf-8 -*-
import os
import sys
import argparse
import numpy as np
import deeptools.countReadsPerBin as countR
from deeptools import parserCommon
from deeptools.utilities import smartLabels
from importlib.metadata import version
old_settings = np.seterr(all='ignore')
def parse_arguments(args=None):
parser = \
argparse.ArgumentParser(
formatter_class=argparse.RawDescriptionHelpFormatter,
description="""
``multiBamSummary`` computes the read coverages for genomic regions for typically two or more BAM files.
The analysis can be performed for the entire genome by running the program in 'bins' mode.
If you want to count the read coverage for specific regions only, use the ``BED-file`` mode instead.
The standard output of ``multiBamSummary`` is a compressed numpy array (``.npz``).
It can be directly used to calculate and visualize pairwise correlation values between the read coverages using the tool 'plotCorrelation'.
Similarly, ``plotPCA`` can be used for principal component analysis of the read coverages using the .npz file.
Note that using a single bigWig file is only recommended if you want to produce a bedGraph file (i.e., with the ``--outRawCounts`` option; the default output file cannot be used by ANY deepTools program if only a single file was supplied!).
A detailed sub-commands help is available by typing:
multiBamSummary bins -h
multiBamSummary BED-file -h
""",
epilog='example usages:\n'
'multiBamSummary bins --bamfiles file1.bam file2.bam -o results.npz \n\n'
'multiBamSummary BED-file --BED selection.bed --bamfiles file1.bam file2.bam \n'
'-o results.npz'
' \n\n',
conflict_handler='resolve')
parser.add_argument('--version', action='version',
version='%(prog)s {}'.format(version('deeptools')))
subparsers = parser.add_subparsers(
title="commands",
dest='command',
description='subcommands',
help='subcommands',
metavar='')
parent_parser = parserCommon.getParentArgParse(binSize=False)
read_options_parser = parserCommon.read_options()
# bins mode options
subparsers.add_parser(
'bins',
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
parents=[bamcorrelate_args(case='bins'),
parent_parser, read_options_parser,
parserCommon.gtf_options(suppress=True)
],
help="The coverage calculation is done for consecutive bins of equal "
"size (10 kilobases by default). This mode is useful to assess the "
"genome-wide similarity of BAM files. The bin size and "
"distance between bins can be adjusted.",
add_help=False,
usage='%(prog)s '
'--bamfiles file1.bam file2.bam '
'-o results.npz \n'
'help: multiBamSummary bins -h / multiBamSummary bins --help\n')
# BED file arguments
subparsers.add_parser(
'BED-file',
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
parents=[bamcorrelate_args(case='BED-file'),
parent_parser, read_options_parser,
parserCommon.gtf_options()
],
help="The user provides a BED file that contains all regions "
"that should be considered for the coverage analysis. A "
"common use is to compare ChIP-seq coverages between two "
"different samples for a set of peak regions.",
usage='%(prog)s --BED selection.bed --bamfiles file1.bam file2.bam -o results.npz\n'
'help: multiBamSummary BED-file -h / multiBamSummary bins --help\n',
add_help=False)
return parser
def bamcorrelate_args(case='bins'):
parser = argparse.ArgumentParser(add_help=False)
required = parser.add_argument_group('Required arguments')
# define the arguments
required.add_argument('--bamfiles', '-b',
metavar='FILE1 FILE2',
help='List of indexed bam files separated by spaces.',
nargs='+',
required=True)
required.add_argument('--outFileName', '-out', '-o',
help='File name to save the coverage matrix. This matrix '
'can be subsequently plotted using plotCorrelation or '
'or plotPCA.',
type=parserCommon.writableFile)
optional = parser.add_argument_group('Optional arguments')
optional.add_argument("--help", "-h", action="help",
help="show this help message and exit")
optional.add_argument('--labels', '-l',
metavar='sample1 sample2',
help='User defined labels instead of default labels from '
'file names. '
'Multiple labels have to be separated by a space, e.g. '
'--labels sample1 sample2 sample3',
nargs='+')
optional.add_argument('--smartLabels',
action='store_true',
help='Instead of manually specifying labels for the input '
'BAM files, this causes deepTools to use the file name '
'after removing the path and extension.')
optional.add_argument('--genomeChunkSize',
type=int,
default=None,
help='Manually specify the size of the genome provided to each processor. '
'The default value of None specifies that this is determined by read '
'density of the BAM file.')
if case == 'bins':
optional.add_argument('--binSize', '-bs',
metavar='INT',
help='Length in bases of the window used '
'to sample the genome. (Default: %(default)s)',
default=10000,
type=int)
optional.add_argument('--distanceBetweenBins', '-n',
metavar='INT',
help='By default, multiBamSummary considers consecutive '
'bins of the specified --binSize. However, to '
'reduce the computation time, a larger distance '
'between bins can by given. Larger distances '
'result in fewer bins considered. (Default: %(default)s)',
default=0,
type=int)
required.add_argument('--BED',
help=argparse.SUPPRESS,
default=None)
else:
optional.add_argument('--binSize', '-bs',
help=argparse.SUPPRESS,
default=10000,
type=int)
optional.add_argument('--distanceBetweenBins', '-n',
help=argparse.SUPPRESS,
metavar='INT',
default=0,
type=int)
required.add_argument('--BED',
help='Limits the coverage analysis to '
'the regions specified in these files.',
metavar='FILE1.bed FILE2.bed',
nargs='+',
required=True)
group = parser.add_argument_group('Output optional options')
group.add_argument('--outRawCounts',
help='Save the counts per region to a tab-delimited file.',
type=parserCommon.writableFile,
metavar='FILE')
group.add_argument('--scalingFactors',
help='Compute scaling factors (in the DESeq2 manner) '
'compatible for use with bamCoverage and write them to a '
'file. The file has tab-separated columns "sample" and '
'"scalingFactor".',
type=parserCommon.writableFile,
metavar='FILE')
return parser
def process_args(args=None):
args = parse_arguments().parse_args(args)
if len(sys.argv) == 1:
parse_arguments().print_help()
sys.exit()
if args.labels and len(args.bamfiles) != len(args.labels):
print("The number of labels does not match the number of bam files.")
exit(0)
if not args.labels:
if args.smartLabels:
args.labels = smartLabels(args.bamfiles)
else:
args.labels = [os.path.basename(x) for x in args.bamfiles]
return args
def main(args=None):
"""
1. get read counts at different positions either
all of same length or from genomic regions from the BED file
2. save data for further plotting
"""
args = process_args(args)
if 'BED' in args:
bed_regions = args.BED
else:
bed_regions = None
if len(args.bamfiles) == 1 and not (args.outRawCounts or args.scalingFactors):
sys.stderr.write("You've input a single BAM file and not specified "
"--outRawCounts or --scalingFactors. The resulting output will NOT be "
"useful with any deepTools program!\n")
stepsize = args.binSize + args.distanceBetweenBins
c = countR.CountReadsPerBin(
args.bamfiles,
args.binSize,
numberOfSamples=None,
genomeChunkSize=args.genomeChunkSize,
numberOfProcessors=args.numberOfProcessors,
verbose=args.verbose,
region=args.region,
bedFile=bed_regions,
blackListFileName=args.blackListFileName,
extendReads=args.extendReads,
minMappingQuality=args.minMappingQuality,
ignoreDuplicates=args.ignoreDuplicates,
center_read=args.centerReads,
samFlag_include=args.samFlagInclude,
samFlag_exclude=args.samFlagExclude,
minFragmentLength=args.minFragmentLength,
maxFragmentLength=args.maxFragmentLength,
stepSize=stepsize,
zerosToNans=False,
out_file_for_raw_data=args.outRawCounts)
num_reads_per_bin = c.run(allArgs=args)
sys.stderr.write("Number of bins "
"found: {}\n".format(num_reads_per_bin.shape[0]))
if num_reads_per_bin.shape[0] < 2:
exit("ERROR: too few non zero bins found.\n"
"If using --region please check that this "
"region is covered by reads.\n")
# numpy will append .npz to the file name if we don't do this...
if args.outFileName:
f = open(args.outFileName, "wb")
np.savez_compressed(f,
matrix=num_reads_per_bin,
labels=args.labels)
f.close()
if args.scalingFactors:
f = open(args.scalingFactors, 'w')
f.write("sample\tscalingFactor\n")
scalingFactors = countR.estimateSizeFactors(num_reads_per_bin)
for sample, scalingFactor in zip(args.labels, scalingFactors):
f.write("{}\t{:6.4f}\n".format(sample, scalingFactor))
f.close()
if args.outRawCounts:
# append to the generated file the
# labels
header = "#'chr'\t'start'\t'end'\t"
header += "'" + "'\t'".join(args.labels) + "'\n"
f = open(args.outRawCounts, 'r+')
content = f.read()
f.seek(0, 0)
f.write(header + content)
f.close()
if __name__ == "__main__":
main()
|
