Mercurial > repos > xuebing > sharplabtool

comparison tools/rgenetics/rgFastQC.py @ 0:9071e359b9a3

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author xuebing
date Fri, 09 Mar 2012 19:37:19 -0500
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:9071e359b9a3
1 """
2 wrapper for fastqc
3
4 called as
5 <command interpreter="python">
6 rgFastqc.py -i $input_file -d $html_file.files_path -o $html_file -n "$out_prefix"
7 </command>
8
9
10
11 Current release seems overly intolerant of sam/bam header strangeness
12 Author notified...
13
14
15 """
16
17 import os,sys,subprocess,optparse,shutil,tempfile
18 from rgutils import getFileString
19
20 class FastQC():
21 """wrapper
22 """
23
24
25 def __init__(self,opts=None):
26 assert opts <> None
27 self.opts = opts
28
29
30 def run_fastqc(self):
31 """
32 In batch mode fastqc behaves not very nicely - will write to a new folder in
33 the same place as the infile called [infilebasename]_fastqc
34 rlazarus@omics:/data/galaxy/test$ ls FC041_1_sequence_fastqc
35 duplication_levels.png fastqc_icon.png per_base_n_content.png per_sequence_gc_content.png summary.txt
36 error.png fastqc_report.html per_base_quality.png per_sequence_quality.png tick.png
37 fastqc_data.txt per_base_gc_content.png per_base_sequence_content.png sequence_length_distribution.png warning.png
38
39 """
40 dummy,tlog = tempfile.mkstemp(prefix='rgFastQClog')
41 sout = open(tlog, 'w')
42 fastq = os.path.basename(self.opts.input)
43 cl = [self.opts.executable,'-o %s' % self.opts.outputdir]
44 if self.opts.informat in ['sam','bam']:
45 cl.append('-f %s' % self.opts.informat)
46 if self.opts.contaminants <> None :
47 cl.append('-c %s' % self.opts.contaminants)
48 cl.append(self.opts.input)
49 p = subprocess.Popen(' '.join(cl), shell=True, stderr=sout, stdout=sout, cwd=self.opts.outputdir)
50 return_value = p.wait()
51 sout.close()
52 runlog = open(tlog,'r').readlines()
53 os.unlink(tlog)
54 flist = os.listdir(self.opts.outputdir) # fastqc plays games with its output directory name. eesh
55 odpath = None
56 for f in flist:
57 d = os.path.join(self.opts.outputdir,f)
58 if os.path.isdir(d):
59 if d.endswith('_fastqc'):
60 odpath = d
61 hpath = None
62 if odpath <> None:
63 try:
64 hpath = os.path.join(odpath,'fastqc_report.html')
65 rep = open(hpath,'r').readlines() # for our new html file but we need to insert our stuff after the <body> tag
66 except:
67 pass
68 if hpath == None:
69 res = ['## odpath=%s: No output found in %s. Output for the run was:<pre>\n' % (odpath,hpath),]
70 res += runlog
71 res += ['</pre>\n',
72 'Please read the above for clues<br/>\n',
73 'If you selected a sam/bam format file, it might not have headers or they may not start with @HD?<br/>\n',
74 'It is also possible that the log shows that fastqc is not installed?<br/>\n',
75 'If that is the case, please tell the relevant Galaxy administrator that it can be snarfed from<br/>\n',
76 'http://www.bioinformatics.bbsrc.ac.uk/projects/fastqc/<br/>\n',]
77 return res
78 self.fix_fastqcimages(odpath)
79 flist = os.listdir(self.opts.outputdir) # these have now been fixed
80 excludefiles = ['tick.png','warning.png','fastqc_icon.png','error.png']
81 flist = [x for x in flist if not x in excludefiles]
82 for i in range(len(rep)): # need to fix links to Icons and Image subdirectories in lastest fastqc code - ugh
83 rep[i] = rep[i].replace('Icons/','')
84 rep[i] = rep[i].replace('Images/','')
85
86 html = self.fix_fastqc(rep,flist,runlog)
87 return html
88
89
90
91 def fix_fastqc(self,rep=[],flist=[],runlog=[]):
92 """ add some of our stuff to the html
93 """
94 bs = '</body></html>\n' # hope they don't change this
95 try:
96 bodyindex = rep.index(bs) # hope they don't change this
97 except:
98 bodyindex = len(rep) - 1
99 res = []
100 res.append('<table>\n')
101 flist.sort()
102 for i,f in enumerate(flist):
103 if not(os.path.isdir(f)):
104 fn = os.path.split(f)[-1]
105 res.append('<tr><td><a href="%s">%s</a></td></tr>\n' % (fn,getFileString(fn, self.opts.outputdir)))
106 res.append('</table><p/>\n')
107 res.append('<a href="http://www.bioinformatics.bbsrc.ac.uk/projects/fastqc/">FastQC documentation and full attribution is here</a><br/><hr/>\n')
108 res.append('FastQC was run by Galaxy using the rgenetics rgFastQC wrapper - see http://rgenetics.org for details and licensing\n')
109 fixed = rep[:bodyindex] + res + rep[bodyindex:]
110 return fixed # with our additions
111
112
113 def fix_fastqcimages(self,odpath):
114 """ Galaxy wants everything in the same files_dir
115 """
116 icpath = os.path.join(odpath,'Icons')
117 impath = os.path.join(odpath,'Images')
118 for adir in [icpath,impath,odpath]:
119 if os.path.exists(adir):
120 flist = os.listdir(adir) # get all files created
121 for f in flist:
122 if not os.path.isdir(os.path.join(adir,f)):
123 sauce = os.path.join(adir,f)
124 dest = os.path.join(self.opts.outputdir,f)
125 shutil.move(sauce,dest)
126 os.rmdir(adir)
127
128
129
130 if __name__ == '__main__':
131 op = optparse.OptionParser()
132 op.add_option('-i', '--input', default=None)
133 op.add_option('-o', '--htmloutput', default=None)
134 op.add_option('-d', '--outputdir', default="/tmp/shortread")
135 op.add_option('-f', '--informat', default='fastq')
136 op.add_option('-n', '--namejob', default='rgFastQC')
137 op.add_option('-c', '--contaminants', default=None)
138 op.add_option('-e', '--executable', default='fastqc')
139 opts, args = op.parse_args()
140 assert opts.input <> None
141 assert os.path.isfile(opts.executable),'##rgFastQC.py error - cannot find executable %s' % opts.executable
142 if not os.path.exists(opts.outputdir):
143 os.makedirs(opts.outputdir)
144 f = FastQC(opts)
145 html = f.run_fastqc()
146 f = open(opts.htmloutput, 'w')
147 f.write(''.join(html))
148 f.close()
149