def tamo2tf(TAMO_file):
'''Converts TAMO files to the TRANSFAC format
'''
ml = MotifTools.txt2motifs(TAMO_file)
TAMO_file_name = TAMO_file.split("/")[-1]
ACGT = ["A", "C", "G", "T"]
n = 1
oup = open("%s.tf" % (TAMO_file), "w")
for m in ml:
if m.source == "":
oup.write("DE\t%s_%s\t%s_%s\n" %
(TAMO_file_name, n, TAMO_file_name, n))
else:
oup.write("DE\t%s\t%s\n" % (m.source, m.source))
count = 0
#print m.source
for i in range(m.width):
oup.write("%s\t" % count)
for letter in ACGT:
if m.logP:
Pij = pow(2.0, m.logP[i][letter])
oup.write("%s\t" % int(Pij * 100))
oup.write("\n")
count += 1
oup.write("XX\n")
n += 1
oup.close()
def parse_opts():
global GLOBALS
global DFUNC, DMAX
short_opts = 'm:'
long_opts = ['dfunc:']
try: opts, args = getopt.getopt(sys.argv[1:], short_opts, long_opts)
except getopt.GetoptError:
print getopt.GetoptError.__dict__
usage()
if not opts: usage()
GLOBALS['args'] = args
GLOBALS['motifs'] = []
DFUNCtxt = None
for opt,value in opts:
if opt == '-m': GLOBALS['motifs'] = MotifTools.txt2motifs(value)
if opt == '--dfunc': DFUNCtxt = value
if opt == '-d': DMAX = float(value)
# Deal with DFUNC and DMAX
if DFUNCtxt == 'NCB':
_DFUNC = MotifCompare.negcommonbits
elif DFUNCtxt:
try:
exec ("_DFUNC = MotifCompare.%s"%DFUNCtxt)
except:
usage("No such distance metric: %s"%DFUNCtxt)
if _DFUNC: set_dfunc(_DFUNC,DMAX)
def pick_chunk_score(wdir, TAMO_file, target, genome):
'''Trims and returns the top motif in a cluster.
This script takes in the TAMO file from the motifs in a single cluster. It
trims the low-information ends from each motifs. It then indentifies the
motif that is most significantly represented in the target genes in your
genome. If no motif is significantly represented, then a blank top motif
file is created.
'''
os.system("cd %s" % wdir)
os.chdir(wdir)
script_dir = '/'.join(os.path.abspath(__file__).split('/')
[:-1]) # path to pcc_merge_CC.py script
##
# step 1 trim tamo to eliminate low information flanking sequence
trim_motif(TAMO_file, 0.1)
##
# step 2 Group Specificity Score" from the Church lab
# python MotifMetrics.py [Genes of interest] -genome [FASTA of promoter sequence] -t [Trimmed TAMO of cluster motifs]
# MotifMetrics.py checks if the motifs appear disproportionatly to the
# targets compared to the rest of the genes.
os.system(
"python %s/MotifMetrics.py %s -genome %s -t %s_0.1.trim -spec > %s_0.1.trim_Cout"
% (script_dir, target, genome, TAMO_file, TAMO_file))
##
# Gets the motif that is most significantly represented in your target genes
# Returns "None" if none of the motifs has a p-value above 0.001.
topm = parse_out_pcs("%s_0.1.trim_Cout" % TAMO_file)
print "topm", topm
##
# Writes the top motif to its own directory.
if topm != "None":
newdic = {}
ml = MotifTools.txt2motifs("%s_0.1.trim" % TAMO_file)
for m in ml:
if m.oneletter == topm:
newdic[m.oneletter] = m
save_motifs(newdic.values(), "%s.TOP" % TAMO_file)
os.system("rm %s_0.1.trim" % TAMO_file)
os.system("rm %s_0.1.trim_Cout" % TAMO_file)
##
# Writes a blank document if there was no top motif.
else:
oup = open("%s.TOP" % TAMO_file, "w")
oup.close()
def TAMO_split(TAMO_file, motifs_per_file=190):
'''This function splits a TAMO into smaller files for create_cc'''
ml = MotifTools.txt2motifs(TAMO_file)
total = len(ml) / int(motifs_per_file) # Total number of TAMOs to generate
by = motifs_per_file
for i in range(total):
print i
print i * by + by, TAMO_file + '_n%s' % i
save_motifs(ml[i * by:i * by + by], TAMO_file + '_n%s' % i)
print total * by, len(ml), TAMO_file + '_n%s' % (total)
save_motifs(ml[total * by:len(ml)], TAMO_file + '_n%s' % (total))
return (total)
def combine_distance_matrix_for_2(wdir, TAMO_file_1, TAMO_file_2):
'''Combines matricies made from two TAMO files.
This script is used to create the final matrix after all jobs from
create_cc_for_2 are complete.
'''
ml_1 = MotifTools.txt2motifs(TAMO_file_1)
ml_2 = MotifTools.txt2motifs(TAMO_file_2)
n_split_1 = len(ml_1) / 100
n_split_2 = len(ml_2) / 100
print n_split_1, len(ml_1)
print n_split_2
# Change to the working directory.
os.system("cd %s" % wdir)
os.chdir(wdir)
# This loop will paste together matricies
for i in range(n_split_1 + 1):
com = "paste "
for j in range(n_split_2 + 1):
com += "%s_n%s-%s_n%s.dm " % (TAMO_file_1, i, TAMO_file_2, j)
com += "> distance_%s" % i
print com
os.system(com)
#
com = "cat "
for i in range(n_split_1 + 1):
com += "distance_%s " % i
com += "> %s-%s.dm" % (TAMO_file_1, TAMO_file_2)
print com
os.system(com)
def parse_opts():
global GLOBALS
short_opts = 'm:g:'
long_opts = ['genome=','top=']
try: opts, args = getopt.getopt(sys.argv[1:], short_opts, long_opts)
except getopt.GetoptError:
print getopt.GetoptError.__dict__
usage()
if not opts: usage()
GLOBALS['args'] = args
for opt,value in opts:
if opt == '-m': GLOBALS['motifs'] = MotifTools.txt2motifs(value)
if opt in ['-g', '--genome']: GLOBALS['genomefile'] = value
if opt == '--top': GLOBALS['top'] = int(value)
def main():
try:
opts, args = getopt.getopt(sys.argv[1:], "f:m:n:L:t:a:S:i:", ["help", "output="]) # AD added 'i'
except getopt.GetoptError:
usage()
sys.exit(1)
if not opts:
usage()
sys.exit(1)
print "#" + ' '.join(sys.argv)
fastafile, motiffile, motifnums, labels, thresh = (None, None, [], None, 0.75) # AD changed thresh val to 0.75 from 0.7
ambigs = []
scale = 50.0 / 1000.0
motifs = []
for opt, value in opts:
#print opt, value
if opt == '-f': fastafile = value
elif opt == '-m': motifs.extend(MotifTools.txt2motifs(value))
elif opt == '-n': motifnums = [int(x) for x in value.split(',')]
elif opt == '-L': labels = list(value)
elif opt == '-t': thresh = float(value)
elif opt == '-a': ambigs.extend(value.split(','))
elif opt == '-S': scale = float(value)
elif opt == '-i': motiffile = value # AD added this option to ACTUALLY supply the tamo motif file at the command-line. The code to deal with motiffiles already existed. There was just no code for User to supply one.
probes = Fasta.load(fastafile)
if motiffile:
for f in motiffile.split(','): # AD added this to allow supplying multiple tamo files at the prompt like you can supply multiple motifs
motifs.extend(MotifTools.load(f))
if ambigs:
for ambig in ambigs:
motifs.append( MotifTools.Motif_from_text(ambig,0.1) )
if not motifnums: motifnums = range(len(motifs))
print '# %d: %s'%(len(motifs),motifnums)
for i in range(len(motifnums)):
motif = motifs[motifnums[i]]
if labels and i < len(labels):
txt = labels[i]
else:
txt = '%d'%i
print '%-3s : %s %5.2f (%4.2f)'%(txt,motif,thresh*motif.maxscore,thresh)
probehits = {}
for key in probes.keys():
hits_by_motif = []
save_flag = 0
if re.search('[BDHU]',probes[key]): continue
for num in motifnums:
result = motifs[num].scan(probes[key],thresh*motif.maxscore)
if result[0]:
hits_by_motif.append(result)
save_flag = 1
else:
hits_by_motif.append(None)
if save_flag:
probehits[key]=hits_by_motif
#scale = .1
maxw = 40
for key in probehits.keys():
l = len(probes[key])
a = list('-'* int(scale*l) )
a.extend( list(' '*10 ) )
desc = []
matches = probehits[key]
for i in range(len(matches)):
if matches[i]:
subseqs,endpoints,scores = matches[i]
for idx in range(len(subseqs)):
start,stop = endpoints[idx]
subseq = subseqs[idx]
score = scores[idx]
if labels and (i<len(labels)): ID = labels[i]
else : ID = '%d'%i
desc.append('%s %s %d-%d %4.2f '%(ID,subseq,start,stop,score))
start = int(start*scale)
for offset in range(10):
if a[start+offset] == '-':
if labels and (i < len(labels)):
a[start+offset] = labels[i]
else:
a[start+offset] = '%d'%i
break
print '%-14s %s'%(key,''.join(a)),
print ' '*max(0,maxw-len(a)), '| '.join(['%-27s'%x for x in desc])
print
print "Found matches in %d of %d input probes"%(len(probehits),len(probes))
def merge_runs_cc(TAMO_file, wdir, height, distance, ancestor, target, genome):
'''This script is used to merge motifs with the PCC matrix of all motifs.
The script was originally written by Cheng Zou, and then converted to a
function by Alex Seddon.
'''
print "Here are the parameters you specified in this run "
print "-tamo %s" % TAMO_file
print "-wdir %s" % wdir
print "-h height to cut the tree, %s" % height
print "-ancestor %s" % ancestor
print "-target %s" % target
print "-genome %s" % genome
if TAMO_file == '' or wdir == '':
help()
os.system("cd %s" % wdir)
os.chdir(wdir)
# Get the directory where the script is located.
script_dir = '/'.join(os.path.abspath(__file__).split('/')[:-1])
# This code was in the original clustering script. It has been taken out
# because the processes involved take too long and have been taken up by
# the matrrix creation scripts and the run_UPGMA script.
#if distance==0:
# os.system("python /mnt/home/seddonal/gil scottscripts/5_motif_merging/3.calculate_distance_matrix.py -i %s --dfunc pccrange" % TAMO_file)
#os.system("R --vanilla --slave --args %s.dm %s< /mnt/home/seddonal/scripts/5_motif_merging/UPGMA_final.R> %s.Rout" % (TAMO_file,height,TAMO_file))
cl_dic = {}
n = 0
# The file, TAMO_file.dm_UPGMA_Cl_0.05, is inorder of the motifs that appear
# in the TAMO_file. If two motifs have the same number, they are considered
# a part of the same cluster.
# This loop pulls the clustering information out of this file and creats
# the dictionary cl_dic = {cluster_index:{motif_index:'1'}}
for line in open("%s.dm_UPGMA_Cl_%s" % (TAMO_file, height), "r"):
# Gets the clusterindex of this motif
cl = line.strip()
# Adds the cluster index if it has not been
if not cl_dic.has_key(cl):
cl_dic[cl] = {}
cl_dic[cl][n] = "1" # Adds the motif to that cluster
n += 1 # Increases the motif index for the next motif
#print cl_dic
ml = MotifTools.txt2motifs(TAMO_file)
old = [] # List of motifs that are the sole members of a cluster.
# I think I can divide up this portion of the code to create a series
print ancestor, ancestor == 0
cc_output = open('merge_runs_cc', 'w')
if ancestor == 0:
# This loop Looks at each cluster and attempts to merge the motifs
# in the cluster if there are multiple motifs.
for i in cl_dic.keys():
print i, cl_dic[i]
# If there are multiple motifs in the cluster, it merges the motifs
if len(cl_dic[i]) > 1:
# Adds all of the motifs in the cluster to an object called
# mlist.
mlist = []
for j in cl_dic[i]:
mlist.append(ml[j])
# Saves these motifs to there own TAMO file.
save_motifs(mlist, "%s_sub_%s.tm" % (TAMO_file, i))
cc_output.write(
'module load TAMO; python %s/pcc_merge_CC.py merge_runs_no_ancestor -t %s/%s -i %s -target %s -genome %s\n'
% (script_dir, wdir, TAMO_file, i, target, genome))
# If there is only one motif in the cluster, it leaves it alone,
# And adds it to old
else:
key = cl_dic[i].keys()[0]
old.append(ml[key])
if ancestor == 1:
# This loop Looks at each cluster and attempts to merge the motifs
# in the cluster if there are multiple motifs.
for i in cl_dic.keys():
print i, cl_dic[i]
# If there are multiple motifs in the cluster, it merges the motifs
if len(cl_dic[i]) > 1:
# Adds all of the motifs in the cluster to an object called
# mlist.
mlist = []
for j in cl_dic[i]:
mlist.append(ml[j])
# Saves these motifs to there own TAMO file.
save_motifs(mlist, "%s_sub_%s.tm" % (TAMO_file, i))
cc_output.write(
'module load TAMO; module load STAMPmotif; python %s/pcc_merge_CC.py merge_runs_ancestor -t %s/%s -i %s -target %s -genome %s\n'
% (script_dir, wdir, TAMO_file, i, target, genome))
else:
key = cl_dic[i].keys()[0]
old.append(ml[key])
# Combine together the motifs that are in there own cluster.
#os.system("cat %s_sub_*_sum.tm.tf.tm.TOP > %s_sub_new.tm" % (TAMO_file,TAMO_file))
save_motifs(old, "%s_sub_old.tm" % (TAMO_file))
def merge_runs(TAMO_file, wdir, height, distance, ancestor, target, genome):
'''This script is used to merge motifs with the PCC matrix of all motifs.
The script was originally written by Cheng Zou, and then converted to a
function by Alex Seddon.
'''
print "Here are the parameters you specified in this run "
print "-tamo %s" % TAMO_file
print "-wdir %s" % wdir
print "-h height to cut the tree, %s" % height
print "-distance %s" % distance
print "-ancestor %s" % ancestor
print "-target %s" % target
print "-genome %s" % genome
if TAMO_file == '' or wdir == '':
help()
os.system("cd %s" % wdir)
os.chdir(wdir)
# This code was in the original clustering script. It has been taken out
# because the processes involved take too long and have been replaced by
# the matrix creation scripts and the run_UPGMA script.
#if distance==0:
# os.system("python /mnt/home/seddonal/scripts/5_motif_merging/3.calculate_distance_matrix.py -i %s --dfunc pccrange" % TAMO_file)
#os.system("R --vanilla --slave --args %s.dm %s< /mnt/home/seddonal/scripts/5_motif_merging/UPGMA_final.R> %s.Rout" % (TAMO_file,height,TAMO_file))
cl_dic = {}
n = 0
# The file, TAMO_file.dm_UPGMA_Cl_0.05, is inorder of the motifs that appear
# in the TAMO_file. If two motifs have the same number, they are considered
# a part of the same cluster.
# This loop pulls the clustering information out of this file and creats
# the dictionary cl_dic = {cluster_index:{motif_index:'1'}}
for line in open("%s.dm_UPGMA_Cl_%s" % (TAMO_file, height), "r"):
# Gets the clusterindex of this motif
cl = line.strip()
# Adds the cluster index if it has not been
if not cl_dic.has_key(cl):
cl_dic[cl] = {}
cl_dic[cl][n] = "1" # Adds the motif to that cluster
n += 1 # Increases the motif index for the next motif
#print cl_dic
ml = MotifTools.txt2motifs(TAMO_file)
old = [] # List of motifs that are the sole members of a cluster.
# I think I can divide up this portion of the code to create a series
print ancestor, ancestor == 0
if ancestor == 0:
# This loop Looks at each cluster and attempts to merge the motifs
# in the cluster if there are multiple motifs.
for i in cl_dic.keys():
print i, cl_dic[i]
# If there are multiple motifs in the cluster, it merges the motifs
if len(cl_dic[i]) > 1:
# Adds all of the motifs in the cluster to an object called
# mlist.
mlist = []
for j in cl_dic[i]:
mlist.append(ml[j])
# Saves these motifs to there own TAMO file.
save_motifs(mlist, "%s_sub_%s.tm" % (TAMO_file, i))
# I am fairly certain that this process of converting to TF and
# then returning it to TAMO format is only for keeping the names
# consistent. I need to verify this suspicion
tamo2tf("%s_sub_%s.tm" % (TAMO_file, i))
os.system("cat %s_sub_%s.tm.tf > %s_sub_%s_sum.tm.tf" %
(TAMO_file, i, TAMO_file, i))
tf2tamo("%s_sub_%s_sum.tm.tf" % (TAMO_file, i))
# Gets the top motif in the cluster.
pick_chunk_score(wdir,
'%s_sub_%s_sum.tm.tf.tm' % (TAMO_file, i),
target, genome)
# Removes the files that were created.
os.system("rm %s_sub_%s_sum.tm.tf.tm" % (TAMO_file, i))
os.system("rm %s_sub_%s_sum.tm.tf" % (TAMO_file, i))
os.system("rm -R %s_sub_%s.tm.tf_ST*" % (TAMO_file, i))
# If there is only one motif in the cluster, it leaves it alone,
# And adds it to old
else:
key = cl_dic[i].keys()[0]
old.append(ml[key])
if ancestor == 1:
# This loop Looks at each cluster and attempts to merge the motifs
# in the cluster if there are multiple motifs.
for i in cl_dic.keys():
print i, cl_dic[i]
# If there are multiple motifs in the cluster, it merges the motifs
if len(cl_dic[i]) > 1:
# Adds all of the motifs in the cluster to an object called
# mlist.
mlist = []
for j in cl_dic[i]:
mlist.append(ml[j])
# Saves these motifs to there own TAMO file.
save_motifs(mlist, "%s_sub_%s.tm" % (TAMO_file, i))
# Merges the motifs in the same cluster using STAMP
tamo2tf("%s_sub_%s.tm" % (TAMO_file, i))
# Gets the JASPER motifs that best match the motifs from within
# the cluster.
os.system(
"STAMP -tf %s_sub_%s.tm.tf -sd /home/chengzou/bin/STAMP/ScoreDists/JaspRand_PCC_SWU.scores \
-go 1000 -ge 1000 -cc PCC -align SWU -out %s_sub_%s.tm.tf_STout -chp > %s_sub_%s.tm.tf_STout.log"
% (TAMO_file, i, TAMO_file, i, TAMO_file, i))
parse_out_STAMP(TAMO_file, i)
# combines the JASPER motifs with the cluster motif and then
# converts them all to one TAMO file
os.system(
"cat %s_sub_%s.tm.tf %s_sub_%s.tm.tf_SToutFBP.txt.mod %s_sub_%s.tm.tf_STout_tree_clusters.txt > %s_sub_%s_sum.tm.tf"
% (TAMO_file, i, TAMO_file, i, TAMO_file, i, TAMO_file, i))
tf2tamo("%s_sub_%s_sum.tm.tf" % (TAMO_file, i))
# Gets the top motif within the TAMO file.
pick_chunk_score(wdir,
'%s_sub_%s_sum.tm.tf.tm' % (TAMO_file, i),
target, genome)
# Removes any files created in the processing.
os.system("rm %s_sub_%s_sum.tm.tf.tm" % (TAMO_file, i))
os.system("rm %s_sub_%s_sum.tm.tf" % (TAMO_file, i))
os.system("rm -R %s_sub_%s.tm.tf_ST*" % (TAMO_file, i))
else:
key = cl_dic[i].keys()[0]
old.append(ml[key])
# Combine together the top motifs from every
os.system("cat %s_sub_*_sum.tm.tf.tm.TOP > %s_sub_new.tm" %
(TAMO_file, TAMO_file))
save_motifs(old, "%s_sub_old.tm" % (TAMO_file))
os.system("cat %s_sub_old.tm %s_sub_new.tm > %s_P1.tm" %
(TAMO_file, TAMO_file, TAMO_file))
def combine_distance_matrix(wdir, TAMO_file):
'''Combines the PCC score matricies and outputs them as a single matrix.
Originaly written by Cheng Zou, and converted to a function by Alex Seddon.
'''
ml = MotifTools.txt2motifs(TAMO_file)
n_split = len(ml) / 100
##
# Change to the working directory.
os.system("cd %s" % wdir)
os.chdir(wdir)
#
##
##
# The following loop keeps counts the number of lines in the each of the
# PCC matricies for a comparison of a TAMO file with itself.
lendic = {} # Dictionary with the length of PCC matricies.
for i in range(n_split + 1):
lendic[i] = line_count("%s_n%s.dm" % (TAMO_file, i))
print lendic
#
##
##
# This loop creates files with blanks. The files are used to ensure that
# the PCC-distance matrix is square. The blank files will be created to take
# the place of files that would have been left blank
for i in range(n_split + 1):
for j in range(0, i):
# open the file to add blanks
oup = open("%s_n%s-%s_n%s.dm" % (TAMO_file, i, TAMO_file, j), "w")
print lendic[j], lendic[i]
list = []
# Add a number of "-" to the list equal to the number of lines in
# the self comparison files.
for y in range(lendic[j]):
list.append("-")
for x in range(lendic[i]):
oup.write("%s\n" % "\t".join(list))
oup.close()
#
##
##
# Creates a copy of the self comparison file so that it can be easily picked
# out by the function.
for i in range(n_split + 1):
os.system("cp %s_n%s.dm %s_n%s-%s_n%s.dm" %
(TAMO_file, i, TAMO_file, i, TAMO_file, i))
#
##
##
# This loop will look at each
for i in range(n_split + 1):
com = "paste "
for j in range(n_split + 1):
com += "%s_n%s-%s_n%s.dm " % (TAMO_file, i, TAMO_file, j)
com += "> distance_%s" % i
print com
os.system(com)
com = "cat "
for i in range(n_split + 1):
com += "distance_%s " % i
com += "> %s.dm" % TAMO_file
print com
# Concatonate all the matricies
os.system(com)
# My embarisingly ad hoc way of removing double tabs
remove_double_tabs("%s.dm" % TAMO_file)
threshold = math.pow(10, -float(sys.argv[2]))
maxthreshold = float(sys.argv[3]) # for strong score, using 0.9*max score
ATbias = float(sys.argv[4]) # 0.33
GCbias = float(sys.argv[5]) # 0.17
seq_file = sys.argv[6] # FASTA file of the sequence
tar_dir = "" # Target directory for the output file
#
##
##
#
for i in range(1, len(sys.argv)):
if sys.argv[i] == "-d":
tar_dir = sys.argv[i + 1].rstrip("/")
print tar_dir
ml = MotifTools.txt2motifs(file)
n = 0
new_list = []
##
# Looks at each motif from the TAMO file. Uses the find function from
# motility to find the sequences with that motif.
for Ikey in range(len(ml)):
#print m.ll
time1 = time.time()
m = ml[Ikey] # Pull out the motif from the motif list.
save_motifs([m], file + '_' + str(Ikey)) # Save the motif as a file.
##
def main():
try:
opts, args = getopt.getopt(sys.argv[1:], "f:m:n:L:t:a:S:", ["help", "output="])
except getopt.GetoptError:
usage()
sys.exit(1)
if not opts:
usage()
sys.exit(1)
print "#" + ' '.join(sys.argv)
fastafile, motiffile, motifnums, labels, thresh = (None, None, [], None, 0.7)
ambigs = []
scale = 50.0 / 1000.0
motifs = []
for opt, value in opts:
#print opt, value
if opt == '-f': fastafile = value
elif opt == '-m': motifs.extend(MotifTools.txt2motifs(value))
elif opt == '-n': motifnums = [int(x) for x in value.split(',')]
elif opt == '-L': labels = list(value)
elif opt == '-t': thresh = float(value)
elif opt == '-a': ambigs.extend(value.split(','))
elif opt == '-S': scale = float(value)
probes = Fasta.load(fastafile)
if motiffile:
motifs.extend(TAMO.tamofile2motifs(motiffile))
if ambigs:
for ambig in ambigs:
motifs.append( MotifTools.Motif_from_text(ambig,0.1) )
if not motifnums: motifnums = range(len(motifs))
print '# %d: %s'%(len(motifs),motifnums)
for i in range(len(motifnums)):
motif = motifs[motifnums[i]]
if labels and i < len(labels):
txt = labels[i]
else:
txt = '%d'%i
print '%-3s : %s %5.2f (%4.2f)'%(txt,motif,thresh*motif.maxscore,thresh)
probehits = {}
for key in probes.keys():
hits_by_motif = []
save_flag = 0
if re.search('[BDHU]',probes[key]): continue
for num in motifnums:
result = motifs[num].scan(probes[key],thresh*motif.maxscore)
if result[0]:
hits_by_motif.append(result)
save_flag = 1
else:
hits_by_motif.append(None)
if save_flag:
probehits[key]=hits_by_motif
#scale = .1
maxw = 40
for key in probehits.keys():
l = len(probes[key])
a = list('-'* int(scale*l) )
a.extend( list(' '*10 ) )
desc = []
matches = probehits[key]
for i in range(len(matches)):
if matches[i]:
subseqs,endpoints,scores = matches[i]
for idx in range(len(subseqs)):
start,stop = endpoints[idx]
subseq = subseqs[idx]
score = scores[idx]
if labels and (i<len(labels)): ID = labels[i]
else : ID = '%d'%i
desc.append('%s %s %d-%d %4.2f '%(ID,subseq,start,stop,score))
start = int(start*scale)
for offset in range(10):
if a[start+offset] == '-':
if labels and (i < len(labels)):
a[start+offset] = labels[i]
else:
a[start+offset] = '%d'%i
break
print '%-14s %s'%(key,''.join(a)),
print ' '*max(0,maxw-len(a)), '| '.join(['%-27s'%x for x in desc])
print
print "Found matches in %d of %d input probes"%(len(probehits),len(probes))