def tamofile2motifs(filename):
FID = open(filename,'r')
lines = FID.readlines()
FID.close()
motifs = []
seedD = {}
seedfile = ''
for i in range(len(lines)):
if lines[i][0:10] == 'Log-odds matrix'[0:10]:
w = len(lines[i+1].split())-1
ll = []
for pos in range(w):
ll.append({})
for j in range(0,4):
toks = lines[i+j+2].split()
L = toks[0][1]
for pos in range(w):
ll[pos][L] = float(toks[pos+1])
m = MotifTools.Motif_from_ll(ll)
motifs.append(m)
if lines[i][0:6] == 'Motif '[0:6]:
toks = lines[i].split()
motifs[-1].nseqs = float(re.sub('[\(\)]','',toks[3]))
motifs[-1].totalbits= float(toks[5])
motifs[-1].MAP = float(toks[7])
motifs[-1].seeddist = float(toks[9])
motifs[-1].seednum = int(toks[10][0:-1])
motifs[-1].pvalue = math.pow(10,-float(toks[12]))
if 'ch:' in toks:
motifs[-1].church = math.pow(10,-float(toks[14]))
if lines[i][0:10] == 'Threshold: '[0:10]:
toks = lines[i].split()
motifs[-1].threshold= float(toks[1])
if lines[i][0:5] == 'Seed '[0:5]:
toks = lines[i].split()
id = int(toks[1][0:-1]) #'10:' -> '10'
seedD[id] = toks[2]
if lines[i][0:7] == 'Source: '[0:7]:
motifs[-1].source = lines[i][7:].strip()
if lines[i][0:6] == 'Gamma: '[0:6]:
motifs[-1].gamma = float(lines[i][6:])
if lines[i][0:6] == 'Evalue: '[0:6]:
motifs[-1].evalue = float(lines[i][7:].strip())
if lines[i].find('Using')>=0 and lines[i].find('as seeds')>=0:
'''#Using all (132) motifs in SLT_081503.seeds as seeds:'''
seedfile = lines[i].split()[-3]
for i in range(len(motifs)):
if seedfile: motifs[i].seedfile = seedfile
seednum = motifs[i].seednum
if seedD.has_key(seednum):
motifs[i].seedtxt = seedD[seednum]
return(motifs)
def motif_matrix(fsa, motif, outfile, genome='mm9'):
if genome == 'hg18':
markov = "/nfs/genomes/human_gp_mar_06/hg18_promoters_3000_1000.markov"
else:
markov = "/nfs/data/cwng/chipseq/hypotheses/Mouse.markov"
#Load motif and background adjust PSSM
m = MotifTools.load(motif)
EM.loadMarkovBackground(markov)
bg = EM.theMarkovBackground.zeroth()
F = Fasta.load(fsa, key_func=lambda x: x)
seqs = F.values()
n_seqs = len(seqs)
n_motifs = len(m)
SCORES = np.zeros((n_motifs, n_seqs), dtype='float')
#SHIFTS=np.zeros((n_motifs,n_seqs))
#out=open(outfile,'w')
for i, M in enumerate(m):
ll = M.logP
EM.loadMarkovBackground(markov)
bg = EM.theMarkovBackground.zeroth()
for pos in ll:
for letter in pos.keys():
pos[letter] = pos[letter] - math.log(
bg[letter]) / math.log(2.0)
AM = MotifTools.Motif_from_ll(ll)
#adj_model = MotifTools.Motif_from_ll(ll)
#adj_model.source = M.source
#pssm = MDsupport.Motif2c_PSSM(adj_model)
#w=pssm.width
#shift=[]
#scores=[]
mi, ma = AM.minscore, AM.maxscore
#F_m={}
#Search every seq for given motif above threshold t and print motif centered results
for j, seq in enumerate(seqs):
seq_fwd = seq.upper()
#seq_rev = str(MotifTools.revcomplement(seq_fwd))[::-1]
#scores_fwd = pssm.score_probe(seq_fwd)
#scores_rev = pssm.score_probe(seq_rev)
#max_score=mi
#max_ind=0
#for ind,s in enumerate(scores_fwd):
# if s> max_score:
# max_score=s
# max_ind=ind
# strand='+'
#for ind,s in enumerate(scores_rev):
# if s> max_score:
# max_score=s
# max_ind=ind
# strand='-'
max_score = AM.bestscore(seq_fwd)
mscore = (max_score - mi) / (ma - mi)
#orig=len(seq_fwd)/2
#bind=max_ind+w//2
#d=abs(orig-bind)
SCORES[i, j] = mscore
#SHIFTS[i,j]=d
#out.write('%1.3f\t'%mscore)
#out.write('\n')
#out.close()
#del F
np.savetxt(outfile, SCORES, fmt='%1.3f')
def main():
##########################################################################################
#THEME.py: THEME module for performing cross-validated hypothesis testing on transcription
#factor binding data.
#Usage: python THEME.py foreground_fasta_file (file path) background_fasta_file (file path)
#hypothesis_index (integer) -fse hypothesis_file (file path) -markov markov_background (file path)
#-motif_file output_file (file path) -cv fold cross-validation (integer)
##########################################################################################
if (len(sys.argv)<4):
print "Usage: THEME.py foreground.fsa background.fsa hypotheses.txt"
sys.exit(1)
fg_file = sys.argv[1] #get fasta file with foreground sequences
bg_file = sys.argv[2] #get fasta file with background sequences
test_indices = sys.argv[3] #colon separated indices into fse file
cv_level = 2 #default 2-fold cross-validation
refine = 1
randomize = 0
beta = 0.0
delta = 0.001
motif_file = 'dummy.out'
dump_categories_to_file = 0
test_family = ''
#read in any command line options
for arg, i in zip(sys.argv,range(len(sys.argv))):
if (arg == '-cv'):
cv_level = int(sys.argv[i+1])
if (arg == '-markov'):
markov_file = sys.argv[i+1]
if (arg == '-fse'):
fse_file = sys.argv[i+1]
if (arg == '-norefine'):
refine = 0
if (arg == '-beta'):
beta = float(sys.argv[i+1])
if (arg == '-delta'):
delta = float(sys.argv[i+1])
if (arg == '-randomization'):
randomize = 1
if (arg == '-motif_file'):
motif_file = sys.argv[i+1]
if (arg == '-dump'):
dump_categories_to_file = 1
if (arg == '-family'):
test_family = family
FH = open(motif_file, 'w')
FH.write("******THEME Motif Output******")
FH.close()
random.seed()
cross_val = THEME(fg_file, bg_file, cv_level, markov_file)
if ((beta>0.0)and(beta<1.0)) : cross_val.beta = beta/(1-beta)
cross_val.delta = delta
cross_val.refine = refine
cross_val.randomize = randomize
cross_val.motif_file = motif_file
if (test_family): cross_val.family = test_family
if (dump_categories_to_file):
cross_val.dump = 1
###################################################################################
#get seed sequences that will be tested
###################################################################################
models = []
fses = MotifTools.load(fse_file)
if (test_indices=='all'):
indices = range(len(fses))
else:
indices = []
ivals = test_indices.split(':')
for v in ivals:
indices.append(int(v))
for i in indices:
ll = fses[i].logP
bg = EM.theMarkovBackground.zeroth()
for pos in ll:
for letter in pos.keys():
pos[letter] = pos[letter] - math.log(bg[letter])/math.log(2.0)
adj_bg_model = MotifTools.Motif_from_ll(ll)
adj_bg_model.source = fses[i].source
models.append(adj_bg_model)
(m, err) = cross_val.run_CV(models)
