def exportMotifDiscoverySequences(infile, outfile):
'''export sequences for motif discovery.
This method requires the _interval tables.
For motif discovery, only the sequences with the highest S/N ratio are supplied.
1. The top *motifs_proportion* intervals sorted by peakval
2. Only a region +/- *motifs_halfwidth* around the peak
3. At least *motifs_min_sequences*. If there are not enough sequences
to start with, all will be used.
4. At most *motifs_max_size* sequences will be output.
'''
track = P.snip(infile, "_intervals.load")
dbhandle = connect()
p = P.substituteParameters(**locals())
nseq = PipelineMotifs.writeSequencesForIntervals(
track,
outfile,
dbhandle,
full=False,
masker=P.asList(p['motifs_masker']),
halfwidth=int(p["motifs_halfwidth"]),
maxsize=int(p["motifs_max_size"]),
proportion=p["motifs_proportion"],
min_sequences=p["motifs_min_sequences"],
num_sequences=p["motifs_num_sequences"],
order=p['motifs_score'])
if nseq == 0:
E.warn("%s: no sequences - meme skipped" % outfile)
P.touch(outfile)
def exportMotifDiscoverySequences( infile, outfile ):
'''export sequences for motif discovery.
This method requires the _interval tables.
For motif discovery, only the sequences with the highest S/N ratio are supplied.
1. The top *motifs_proportion* intervals sorted by peakval
2. Only a region +/- *motifs_halfwidth* around the peak
3. At least *motifs_min_sequences*. If there are not enough sequences
to start with, all will be used.
4. At most *motifs_max_size* sequences will be output.
'''
track = P.snip( infile, "_intervals.load" )
dbhandle = connect()
p = P.substituteParameters( **locals() )
nseq = PipelineMotifs.writeSequencesForIntervals( track,
outfile,
dbhandle,
full = False,
masker = P.asList(p['motifs_masker']),
halfwidth = int(p["motifs_halfwidth"]),
maxsize = int(p["motifs_max_size"]),
proportion = p["motifs_proportion"],
min_sequences = p["motifs_min_sequences"],
num_sequences = p["motifs_num_sequences"],
order = p['motifs_score'])
if nseq == 0:
E.warn( "%s: no sequences - meme skipped" % outfile)
P.touch( outfile )
def mapReadsAgainstSpadesContigs(infiles, outfile):
'''
map reads against spades contigs
'''
inf = infiles[0]
to_cluster = True
index_dir = os.path.dirname(outfile)
if "agg" not in infiles[1]:
genome = re.search(
".*R[0-9]*", infiles[0]).group(0) + ".filtered.contigs.fa"
else:
genome = "agg-agg-agg.filtered.contigs.fa"
if infiles[1].endswith(".bt2") or infiles[1].endswith(".ebwt"):
infile, reffile = infiles[0], os.path.join(index_dir, genome) + ".fa"
m = PipelineMapping.Bowtie(
executable=P.substituteParameters(**locals())["bowtie_executable"])
elif infiles[1].endswith("bwt"):
genome = genome
job_options = " -l mem_free=%s" % (PARAMS["bwa_memory"])
bwa_index_dir = index_dir
bwa_mem_options = PARAMS["bwa_mem_options"]
bwa_threads = PARAMS["bwa_threads"]
m = PipelineMapping.BWAMEM(remove_non_unique=True)
statement = m.build((inf,), outfile)
P.run()
def mapReadsWithBowtieAgainstRayContigs(infile, outfile):
'''
map reads against contigs with bowtie
'''
PARAMS["bowtie_index_dir"] = "ray.dir"
PARAMS["genome"] = TRACKS.getTracks(infile)[0].split(".")[0]
infile, reffile = infile, os.path.join("ray.dir", TRACKS.getTracks(infile)[0])
m = PipelineMapping.Bowtie( executable = P.substituteParameters( **locals() )["bowtie_executable"] )
statement = m.build( (infile,), outfile )
P.run()
def mapReadsWithBowtieAgainstRayContigs(infile, outfile):
'''
map reads against contigs with bowtie
'''
PARAMS["bowtie_index_dir"] = "ray.dir"
PARAMS["genome"] = TRACKS.getTracks(infile)[0].split(".")[0]
infile, reffile = infile, os.path.join("ray.dir",
TRACKS.getTracks(infile)[0])
m = PipelineMapping.Bowtie(executable=P.substituteParameters(
**locals())["bowtie_executable"])
statement = m.build((infile, ), outfile)
P.run()
def mapReadsWithBowtieAgainstExpectedContigs(infiles, outfile):
'''
map reads against contigs with bowtie
'''
to_cluster = True
bowtie_index_dir = "expected_contigs.dir"
genome = os.path.basename(re.search(".*R[0-9]*", infiles[0]).group(0) + ".contigs.expected")
for seq in infiles[1]:
to_cluster = True
infile, reffile = seq, genome + ".fa"
m = PipelineMapping.Bowtie( executable = P.substituteParameters( **locals() )["bowtie_executable"] )
statement = m.build( (infile,), outfile )
P.run()
