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 run_mapping(infile, outfile):
''' Map reads with the specified read mapper '''
if PARAMS["mapper"] == "star":
job_threads = PARAMS["star_threads"]
job_memory = PARAMS["star_memory"]
star_mapping_genome = PARAMS["star_genome"] or PARAMS["genome"]
m = PipelineMapping.STAR(
executable=P.substituteParameters(**locals())["star_executable"],
strip_sequence=0)
elif PARAMS["mapper"] == "bowtie":
job_threads = PARAMS["bowtie_threads"]
job_memory = PARAMS["bowtie_memory"]
m = PipelineMapping.Bowtie(executable="bowtie",
tool_options=PARAMS["bowtie_options"],
strip_sequence=0)
genome = PARAMS["bowtie_genome"]
reffile = os.path.join(PARAMS["bowtie_index_dir"],
PARAMS["bowtie_genome"] + ".fa")
statement = m.build((infile, ), outfile)
P.run()
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 mapReadsWithHisat(infiles, outfile):
'''
Map reads using Hisat (spliced reads).
Parameters
----------
infiles: list
contains two filenames -
infiles[0]: str
filename of reads file
can be :term:`fastq`, :term:`sra`, csfasta
infiles[1]: str
filename with suffix .junctions containing a list of known
splice junctions.
hisat_threads: int
:term:`PARAMS`
number of threads with which to run hisat
hisat_memory: str
:term:`PARAMS`
memory required for hisat job
hisat_executable: str
:term:`PARAMS`
path to hisat executable
hisat_library_type: str
:term:`PARAMS`
hisat rna-strandess parameter, see
https://ccb.jhu.edu/software/hisat/manual.shtml#command-line
hisat_options: str
options string for hisat, see
https://ccb.jhu.edu/software/hisat/manual.shtml#command-line
hisat_index_dir: str
path to directory containing hisat indices
strip_sequence: bool
:term:`PARAMS`
if set, strip read sequence and quality information
outfile: str
:term:`bam` filename to write the mapped reads in bam format.
.. note::
If hisat fails with an error such as::
Error: segment-based junction search failed with err =-6
what(): std::bad_alloc
it means that it ran out of memory.
'''
job_threads = PARAMS["hisat_threads"]
job_memory = PARAMS["hisat_memory"]
m = PipelineMapping.Hisat(
executable=P.substituteParameters(
**locals())["hisat_executable"],
strip_sequence=PARAMS["strip_sequence"])
infile, junctions = infiles
infile = P.snip(infile, ".subset") + ".fastq.gz"
if not os.path.exists(infile):
infile = P.snip(infile, ".fastq.gz") + ".fastq.1.gz"
statement = m.build((infile,), outfile)
P.run()
