def generatePSP(positives, negatives, outfile):
''' generate a discrimitative PSP file from
the positives and negatives that can be used
to do descriminative MEME '''
psp_options = PARAMS["psp_options"]
nseqs_pos = int(FastaIterator.count(positives))
nseqs_neg = int(FastaIterator.count(negatives))
if nseqs_pos < 2 or nseqs_neg < 2:
E.warn("%s: input files do not have sufficent sequences"
"to run psp-gen, skipping" % outfile)
P.touch(outfile)
return
# get appropriate options from meme options
if PARAMS.get("meme_revcomp", True):
psp_options += " -revcomp"
statement = '''psp-gen -pos %(positives)s
-neg %(negatives)s
%(psp_options)s
> %(outfile)s '''
P.run(statement)
def runDREME(infile, outfile, neg_file="", options=""):
''' Run DREME on fasta file. If a neg_file is passed
then DREME will use this as the negative set, otherwise
the default is to shuffle the input '''
nseqs_pos = int(FastaIterator.count(infile))
if nseqs_pos < 2:
E.warn("%s: less than 2 sequences - dreme skipped" % outfile)
P.touch(outfile)
return
if neg_file:
nseqs_neg = int(FastaIterator.count(neg_file))
if nseqs_neg < 2:
E.warn(
"%s: less than 2 sequences in negatives file - dreme skipped" %
outfile)
P.touch(outfile)
return
else:
neg_file = "-n %s" % neg_file
logfile = outfile + ".log"
target_path = os.path.join(os.path.abspath(PARAMS["exportdir"]), outfile)
tmpdir = P.get_temp_dir(".")
statement = '''
dreme -p %(infile)s %(neg_file)s -png
-oc %(tmpdir)s
%(dreme_options)s
%(options)s
> %(logfile)s
'''
P.run(statement)
collectMEMEResults(tmpdir, target_path, outfile, method="dreme")
def runMemeCHIP(infile, outfile, motifs=None):
'''Run the MEME-CHiP pipeline on the input files.
optional motifs files can be supplied as a list'''
if motifs:
motifs = " ".join("-db %s" % motif for motif in motifs)
else:
motifs = " "
nseqs = int(FastaIterator.count(infile))
if nseqs == 0:
E.warn("%s: no sequences - meme-chip skipped")
P.touch(outfile)
return
target_path = os.path.join(
os.path.abspath(PARAMS["exportdir"]), outfile)
tmpdir = P.getTempDir(".")
statement = '''
meme-chip %(infile)s
-p %(meme_threads)s
-oc %(tmpdir)s
-nmeme %(memechip_nmeme)s
%(memechip_options)s
%(motifs)s > %(outfile)s.log '''
# If running with more than one thread
# http://git.net/ml/clustering.gridengine.users/2007-04/msg00058.html
# specify "excl=false -w n -pe openmpi-ib num_threads" in cluster_options
# through job_options
if int(PARAMS["memechip_threads"]) != 1:
job_options = str(PARAMS["memechip_job_options"])
job_threads = int(PARAMS["memechip_threads"])
cluster_parallel_environment = str(PARAMS["memechip_cluster_parallel_environment"])
P.run()
collectMEMEResults(tmpdir, target_path, outfile, method="memechip")
def runMEMEOnSequences(infile, outfile):
'''run MEME to find motifs.
In order to increase the signal/noise ratio,
MEME is not run on all intervals but only the
top 10% of intervals (peakval) are used.
Also, only the segment of 200 bp around the peak
is used and not the complete interval.
* Softmasked sequence is converted to hardmasked
sequence to avoid the detection of spurious motifs.
* Sequence is run through dustmasker
'''
to_cluster = True
# job_options = "-l mem_free=8000M"
nseqs = int(FastaIterator.count(infile))
if nseqs == 0:
E.warn("%s: no sequences - meme skipped" % outfile)
P.touch(outfile)
return
target_path = os.path.join(
os.path.abspath(PARAMS["exportdir"]), "meme", outfile)
tmpdir = P.getTempDir(".")
statement = '''
meme %(infile)s -dna -revcomp
-mod %(meme_model)s
-nmotifs %(meme_nmotifs)s
-oc %(tmpdir)s
-maxsize %(motifs_max_size)s
%(meme_options)s
> %(outfile)s.log
'''
P.run()
collectMEMEResults(tmpdir, target_path, outfile)
def runMEMEOnSequences(infile, outfile):
'''run MEME to find motifs.
In order to increase the signal/noise ratio,
MEME is not run on all intervals but only the
top 10% of intervals (peakval) are used.
Also, only the segment of 200 bp around the peak
is used and not the complete interval.
* Softmasked sequence is converted to hardmasked
sequence to avoid the detection of spurious motifs.
* Sequence is run through dustmasker
'''
to_cluster = True
# job_options = "-l mem_free=8000M"
nseqs = int(FastaIterator.count(infile))
if nseqs == 0:
E.warn("%s: no sequences - meme skipped" % outfile)
P.touch(outfile)
return
target_path = os.path.join(os.path.abspath(PARAMS["exportdir"]), "meme",
outfile)
tmpdir = P.getTempDir(".")
statement = '''
meme %(infile)s -dna -revcomp
-mod %(meme_model)s
-nmotifs %(meme_nmotifs)s
-oc %(tmpdir)s
-maxsize %(motifs_max_size)s
%(meme_options)s
> %(outfile)s.log
'''
P.run()
collectMEMEResults(tmpdir, target_path, outfile)
def subsampleNReadsFromFasta(infile, outfile, nreads, logfile=""):
checkParams()
nseqs = FastaIterator.count(infile)
if nreads > nseqs:
prop = 1
else:
prop = float(nreads) / float(nseqs)
if logfile:
logfile = "-L %s" % logfile
statement = ''' python %(scriptsdir)s/fasta2fasta.py
-I %(infile)s
%(logfile)s
-m sample
--sample-proportion=%(prop)s
-S %(outfile)s '''
P.run()
def subsampleNReadsFromFasta(infile, outfile, nreads, logfile=""):
checkParams()
nseqs = FastaIterator.count(infile)
if nreads > nseqs:
prop = 1
else:
prop = float(nreads)/float(nseqs)
if logfile:
logfile = "-L %s" % logfile
statement = ''' python %(scriptsdir)s/fasta2fasta.py
-I %(infile)s
%(logfile)s
-m sample
--sample-proportion=%(prop)s
-S %(outfile)s '''
P.run()
def runMEMEOnSequences(infile, outfile, background=None, psp=None):
'''run MEME on fasta sequences to find motifs
By defualt MEME calculates a zero-th order background
model from the nucleotide frequencies in the input set.
To use a different background set, a background
file created by fasta-get-markov must be supplied.
To perform descrimantive analysis a position specific
prior (psp) file must be provided. This can be generated
used generatePSP.
'''
# job_options = "-l mem_free=8000M"
nseqs = int(FastaIterator.count(infile))
if nseqs < 2:
E.warn("%s: less than 2 sequences - meme skipped" % outfile)
P.touch(outfile)
return
# Get the total length of the sequences to decide the memory
total_seqs_length = 0
with IOTools.open_file(infile, "r") as fasta_reader:
iterator_fasta = FastaIterator.iterate(fasta_reader)
for fasta_seq in iterator_fasta:
total_seqs_length += len(fasta_seq.sequence)
fasta_reader.close()
# If the length of all sequences is higher than 160,000bp
# Up the memory
job_memory = "2G"
if (total_seqs_length > 160000):
job_memory = "4G"
if PARAMS.get("meme_revcomp", True):
revcomp = "-revcomp"
else:
revcomp = ""
target_path = os.path.join(os.path.abspath(PARAMS["exportdir"]), outfile)
tmpdir = P.get_temp_dir(".")
if background:
background_model = "-bfile %s" % background
else:
background_model = ""
if psp:
E.info("Running MEME in descriminative mode")
psp_file = "-psp %s" % psp
else:
psp_file = ""
statement = '''
meme %(infile)s -dna %(revcomp)s
-p %(meme_threads)s
-mod %(meme_model)s
-nmotifs %(meme_nmotifs)s
-oc %(tmpdir)s
-maxsize %(meme_max_size)s
%(background_model)s
%(psp_file)s
%(meme_options)s
2> %(outfile)s.log
'''
# If running with more than one thread
# http://git.net/ml/clustering.gridengine.users/2007-04/msg00058.html
# specify "excl=false -w n -pe openmpi-ib num_threads" in cluster_options
# through job_options
if int(PARAMS["meme_threads"]) != 1:
job_options = str(PARAMS["meme_job_options"])
job_threads = int(PARAMS["meme_threads"])
cluster_parallel_environment = str(
PARAMS["meme_cluster_parallel_environment"])
P.run(statement)
collectMEMEResults(tmpdir, target_path, outfile)
