def main():
usage = "usage: %prog [options] <input FASTA>"
description = "The program applies a modified EXTREME algorithm to find motifs in a FASTA file. It accepts a positive sequence set, a negative sequence set, a list of seed PFMs, and an index number indicating which of the seed PFMs to use"
parser = ArgumentParser(description=description)
parser.add_argument('fastafile', metavar='f', help='FASTA file containing the sequences')
parser.add_argument('negfastafile', metavar='g', help='Negative FASTA file. This is for comparison so that you know the motif you discovered is over-represented.')
parser.add_argument('jfile', metavar='j', help='File containing PWM seeds')
parser.add_argument('indexvalue', metavar='i', help='Which seed from the Minimal MEME Format file to use (it is an integer ranging from 1 to the total number of PFM seeds in your file)', type=int)
parser.add_argument("-p", "--pseudocounts", help="Pseudo counts added to initial PFM guess. Default:0.0", type=float, default=0.0)
parser.add_argument("-maxsites", dest="maxsites", help="Maximum number of expected sites for the motif. If not specified, defaults to 5 times number of initial predicted sites.", type=int, default=0)
parser.add_argument("-minsites", dest="minsites", help="Minimum number of expected sites for the motif. Default: 10", type=int, default=10)
parser.add_argument("-t", "--tries", dest="tries", help="Number of tries for each motif discovered. The fudge factor is changed until the number of discovered sites is in the \"acceptable\" range", type=int, default=15)
parser.add_argument("-s", "--seed", dest="seed", help="Random seed", type=int, default=1)
parser.add_argument("-saveseqs", "--saveseqs", dest="saveseqs", help="If specified, save sequences to current directory", action='store_true')
import time
print "Started at:"
print time.ctime()
starttime = time.time()
args = parser.parse_args()
seed = args.seed
minsites = args.minsites
maxsites = args.maxsites
random.seed(seed)
jfile = open(args.jfile,'r')
from numpy import fromstring
from string import join
lines = jfile.readlines()
j = 0
for i in range(len(lines)):
line = lines[i]
if '>' in line:#This is a name line, so read in next lines for matrix
j += 1
if j == args.indexvalue:#at the desired index
parts = lines[i].split()
pos_cs = parts[1]
motifname = parts[0][1:]
w = len(pos_cs)
strlines = lines[i+1:i+1+w]
pwm_string = ''
for strline in strlines:
strparts = strline.split()
for strpart in strparts:
pwm_string += strpart + ' '
#print pwm_string
pwm_guess = fromstring(pwm_string,sep=' ',dtype=float)
pwm_guess = pwm_guess.reshape((w,4))
break
print 'Using initial motif guess',motifname
print 'Adding',str(args.pseudocounts),'pseudocounts and normalizing'
pwm_guess = pwm_guess + args.pseudocounts
pwm_guess = pwm_guess/pwm_guess.sum(axis=1)[:,newaxis]
jfile.close()
# make the directory (recursively)
import os
outdir = motifname
outpre = outdir + "/"
clobber = True
try:#adapted from DREME.py by T. Bailey
os.makedirs(outdir)
except OSError as exc:
if exc.errno == errno.EEXIST:
if not clobber:
print >> sys.stderr, ("output directory (%s) already exists "
"but EXTREME was not told to clobber it") % (outdir); sys.exit(1)
else: raise
#Use DREME's SeqIO to read in FASTA to list
seqs = sequence.convert_ambigs(sequence.readFASTA(args.fastafile, None, True))
#print seqs
negseqs = sequence.convert_ambigs(sequence.readFASTA(args.negfastafile, None, True))
tries = args.tries
theta_motifs, theta_background_matrices, lambda_motifs, logevs, disc_pwms, disc_logevs, disc_nsites = extreme(seqs,negseqs,minsites,maxsites,pwm_guess,tries)
k = 1
outputMEMEformat(disc_pwms, disc_logevs, disc_nsites, outpre)
try:
from weblogolib import LogoData, LogoOptions, LogoFormat, png_formatter, eps_formatter, unambiguous_dna_alphabet
for theta_motif, theta_background_matrix, lambda_motif, logev in zip(theta_motifs, theta_background_matrices, lambda_motifs, logevs):
outputMotif(theta_motif, theta_background_matrix, lambda_motif, logev, k, outpre)
k = k+1
except ImportError:
print "You do not have Weblogolib, so sequence logos will not be made"
if args.saveseqs:
print "Saving Positive sequences to Positive_seq.fa"
pos_file = open("Positive_seq.fa","w")
for s in range(len(seqs)):
pos_file.write(">sequence"+str(s+1)+"\n")
pos_file.write(seqs[s]+"\n")
pos_file.close()
print "Saving Negative sequences to Negative_seq.fa"
neg_file = open("Negative_seq.fa","w")
for s in range(len(negseqs)):
neg_file.write(">sequence"+str(s+1)+"\n")
neg_file.write(negseqs[s]+"\n")
neg_file.close()
print "Ended at:"
print time.ctime()
stoptime = time.time()
duration = stoptime - starttime
print "Duration:", duration
def main():
usage = "usage: %prog [options] <input FASTA> <negative FASTA>"
description = "The program performs a DREME-like search for gapped k-mers"
parser = ArgumentParser(description=description)
parser.add_argument('fastafile',
metavar='f',
help='FASTA file containing the sequences')
parser.add_argument('negativefile',
metavar='n',
help='FASTA file containing the negative sequences')
parser.add_argument('outputfile', metavar='o', help='Output file')
parser.add_argument(
"-w",
"--width",
dest="width",
help=
"Width of the motif to search for. This makes the program only search for a motif of this width. Beware if greater than 8",
type=int,
default=0)
parser.add_argument("-ming",
dest="mingap",
help="Minimum gap of k-mer to search for. Default: 0",
type=int,
default=0)
parser.add_argument("-maxg",
dest="maxgap",
help="Maximum gap of k-mer to search for. Default: 12",
type=int,
default=10)
parser.add_argument(
"-l",
dest="halflength",
help=
"Number of non-degenerate letters per half-site. Total number of non-degenerate letters is twice this number. Default: 4",
type=int,
default=4)
parser.add_argument(
"-minw",
dest="minwidth",
help=
"Minimum width of the motif to search for. The default is 3, which is the width of the smallest core motif.",
type=int,
default=3)
parser.add_argument(
"-maxw",
dest="maxwidth",
help=
"Maximum width of the motif to search for. This program does one refinement at this width (if greater than 8), and then picks the most significant short-mer. Default: 8",
type=int,
default=8)
parser.add_argument(
"-mink",
dest="mink",
help=
"Minimum width of the core to search for. The default is 3, which is the width of the smallest core motif.",
type=int,
default=3)
parser.add_argument(
"-maxk",
dest="maxk",
help="Maximum width of the core to search for. Default: 8",
type=int,
default=8)
parser.add_argument("-z",
"--zthresh",
dest="zthresh",
help="Corrected z-score threshold. Default: 5",
type=float,
default=5)
parser.add_argument(
"-minsites",
"--minsites",
dest="minsites",
help="Minimum number of sites for a k-mer to be included. Default: 10",
type=int,
default=10)
args = parser.parse_args()
pos_seq_file_name = args.fastafile
neg_seq_file_name = args.negativefile
print 'Reading positive sequence file...'
pos_seqs = sequence.convert_ambigs(
sequence.readFASTA(pos_seq_file_name, None, True))
print 'Reading negative sequence file...'
neg_seqs = sequence.convert_ambigs(
sequence.readFASTA(neg_seq_file_name, None, True))
halflength = args.halflength
ming = args.mingap
maxg = args.maxgap
zthresh = args.zthresh
minsites = args.minsites
find_kmers(pos_seqs, neg_seqs, halflength, ming, maxg, minsites, zthresh,
args.outputfile)
def main():
usage = "usage: %prog [options] <input FASTA> <negative FASTA>"
description = "The program performs a DREME-like search for gapped k-mers"
parser = ArgumentParser(description=description)
parser.add_argument("fastafile", metavar="f", help="FASTA file containing the sequences")
parser.add_argument("negativefile", metavar="n", help="FASTA file containing the negative sequences")
parser.add_argument("outputfile", metavar="o", help="Output file")
parser.add_argument(
"-w",
"--width",
dest="width",
help="Width of the motif to search for. This makes the program only search for a motif of this width. Beware if greater than 8",
type=int,
default=0,
)
parser.add_argument(
"-ming", dest="mingap", help="Minimum gap of k-mer to search for. Default: 0", type=int, default=0
)
parser.add_argument(
"-maxg", dest="maxgap", help="Maximum gap of k-mer to search for. Default: 12", type=int, default=10
)
parser.add_argument(
"-l",
dest="halflength",
help="Number of non-degenerate letters per half-site. Total number of non-degenerate letters is twice this number. Default: 4",
type=int,
default=4,
)
parser.add_argument(
"-minw",
dest="minwidth",
help="Minimum width of the motif to search for. The default is 3, which is the width of the smallest core motif.",
type=int,
default=3,
)
parser.add_argument(
"-maxw",
dest="maxwidth",
help="Maximum width of the motif to search for. This program does one refinement at this width (if greater than 8), and then picks the most significant short-mer. Default: 8",
type=int,
default=8,
)
parser.add_argument(
"-mink",
dest="mink",
help="Minimum width of the core to search for. The default is 3, which is the width of the smallest core motif.",
type=int,
default=3,
)
parser.add_argument(
"-maxk", dest="maxk", help="Maximum width of the core to search for. Default: 8", type=int, default=8
)
parser.add_argument(
"-z", "--zthresh", dest="zthresh", help="Corrected z-score threshold. Default: 5", type=float, default=5
)
parser.add_argument(
"-minsites",
"--minsites",
dest="minsites",
help="Minimum number of sites for a k-mer to be included. Default: 10",
type=int,
default=10,
)
args = parser.parse_args()
pos_seq_file_name = args.fastafile
neg_seq_file_name = args.negativefile
print "Reading positive sequence file..."
pos_seqs = sequence.convert_ambigs(sequence.readFASTA(pos_seq_file_name, None, True))
print "Reading negative sequence file..."
neg_seqs = sequence.convert_ambigs(sequence.readFASTA(neg_seq_file_name, None, True))
halflength = args.halflength
ming = args.mingap
maxg = args.maxgap
zthresh = args.zthresh
minsites = args.minsites
find_kmers(pos_seqs, neg_seqs, halflength, ming, maxg, minsites, zthresh, args.outputfile)
def main():
usage = "usage: %prog [options] <input FASTA>"
description = "The program applies a modified EXTREME algorithm to find motifs in a FASTA file. It accepts a positive sequence set, a negative sequence set, a list of seed PFMs, and an index number indicating which of the seed PFMs to use"
parser = ArgumentParser(description=description)
parser.add_argument('fastafile', metavar='f', help='FASTA file containing the sequences')
parser.add_argument('negfastafile', metavar='g', help='Negative FASTA file. This is for comparison so that you know the motif you discovered is over-represented.')
parser.add_argument('jfile', metavar='j', help='File containing PWM seeds')
parser.add_argument('indexvalue', metavar='i', help='Which seed from the Minimal MEME Format file to use (it is an integer ranging from 1 to the total number of PFM seeds in your file)', type=int)
parser.add_argument("-p", "--pseudocounts", help="Pseudo counts added to initial PFM guess. Default:0.0", type=float, default=0.0)
parser.add_argument("-q", "--initialstep", help="The initial step size for the online EM algorithm. A VERY sensitive parameter. I get best success for ChIP size data (about 100,000 to 1,000,000 bps) with a step size of 0.05. For DNase footprinting, which usually has >5,000,000 bps, I find 0.02 works best. Default:0.05", type=float, default=0.05)
parser.add_argument("-maxsites", dest="maxsites", help="Maximum number of expected sites for the motif. If not specified, defaults to 5 times number of initial predicted sites.", type=int, default=0)
parser.add_argument("-minsites", dest="minsites", help="Minimum number of expected sites for the motif. Default: 10", type=int, default=10)
parser.add_argument("-t", "--tries", dest="tries", help="Number of tries for each motif discovered. The fudge factor is changed until the number of discovered sites is in the \"acceptable\" range", type=int, default=15)
parser.add_argument("-s", "--seed", dest="seed", help="Random seed", type=int, default=1)
parser.add_argument("-saveseqs", "--saveseqs", dest="saveseqs", help="If specified, save sequences to current directory", action='store_true')
parser.add_argument("-b", "--background", dest="background", help="If specified, the minimal MEME output will use the calculated background probabilities instead of uniform probabilities.", action='store_true')
import time
print "Started at:"
print time.ctime()
starttime = time.time()
args = parser.parse_args()
seed = args.seed
initialstep = args.initialstep
minsites = args.minsites
maxsites = args.maxsites
random.seed(seed)
jfile = open(args.jfile,'r')
from numpy import fromstring
from string import join
lines = jfile.readlines()
j = 0
for i in range(len(lines)):
line = lines[i]
if '>' in line:#This is a name line, so read in next lines for matrix
j += 1
if j == args.indexvalue:#at the desired index
parts = lines[i].split()
pos_cs = parts[1]
motifname = parts[0][1:]
w = len(pos_cs)
strlines = lines[i+1:i+1+w]
pwm_string = ''
for strline in strlines:
strparts = strline.split()
for strpart in strparts:
pwm_string += strpart + ' '
#print pwm_string
pwm_guess = fromstring(pwm_string,sep=' ',dtype=float)
pwm_guess = pwm_guess.reshape((w,4))
break
print 'Using initial motif guess',motifname
print 'Adding',str(args.pseudocounts),'pseudocounts and normalizing'
pwm_guess = pwm_guess + args.pseudocounts
pwm_guess = pwm_guess/pwm_guess.sum(axis=1)[:,newaxis]
jfile.close()
# make the directory (recursively)
import os
outdir = motifname
outpre = outdir + "/"
clobber = True
try:#adapted from DREME.py by T. Bailey
os.makedirs(outdir)
except OSError as exc:
if exc.errno == errno.EEXIST:
if not clobber:
print >> sys.stderr, ("output directory (%s) already exists "
"but EXTREME was not told to clobber it") % (outdir); sys.exit(1)
else: raise
#Use DREME's SeqIO to read in FASTA to list
seqs = sequence.convert_ambigs(sequence.readFASTA(args.fastafile, None, True))
#print seqs
negseqs = sequence.convert_ambigs(sequence.readFASTA(args.negfastafile, None, True))
tries = args.tries
theta_motifs, theta_background_matrices, lambda_motifs, logevs, disc_pwms, disc_bkg, disc_logevs, disc_nsites = extreme(seqs,negseqs,minsites,maxsites,pwm_guess,initialstep,tries)
k = 1
outputMEMEformat(disc_pwms, disc_bkg, disc_logevs, disc_nsites, outpre, args.background)
try:
from weblogolib import LogoData, LogoOptions, LogoFormat, png_formatter, eps_formatter, unambiguous_dna_alphabet
for theta_motif, theta_background_matrix, lambda_motif, logev in zip(theta_motifs, theta_background_matrices, lambda_motifs, logevs):
outputMotif(theta_motif, theta_background_matrix, lambda_motif, logev, k, outpre)
k = k+1
except ImportError:
print "You do not have Weblogolib, so sequence logos will not be made"
if args.saveseqs:
print "Saving Positive sequences to Positive_seq.fa"
pos_file = open("Positive_seq.fa","w")
for s in range(len(seqs)):
pos_file.write(">sequence"+str(s+1)+"\n")
pos_file.write(seqs[s]+"\n")
pos_file.close()
print "Saving Negative sequences to Negative_seq.fa"
neg_file = open("Negative_seq.fa","w")
for s in range(len(negseqs)):
neg_file.write(">sequence"+str(s+1)+"\n")
neg_file.write(negseqs[s]+"\n")
neg_file.close()
print "Ended at:"
print time.ctime()
stoptime = time.time()
duration = stoptime - starttime
print "Duration:", duration