from gusPyCode.MDAP_proj.MDAP_defs import findBestPairAlignments
from TAMO.MotifTools import Motif,load
iFiles = ['/Users/biggus/Documents/James/Collaborations/Campbell/data/Results_HyperGeoScreen/masked/Results_gGEMS/CCupAt4Days.6-8mers.gGEMS.top6.motifs.stdThresh.tmo']
motifs = []
for i in range(len(iFiles)):
motifs.extend(load(iFiles[i]))
mat = findBestPairAlignments(motifs, 1)
None
from TAMO.MotifTools import load,sum
import gusPyCode.MDAP_proj.MDAP_defs as md
motifs = load('/Users/biggus/Documents/James/Collaborations/Campbell/data/Results_HyperGeoScreen/masked/Results_gGEMS/CCupAt4Days.6-8mers.gGEMS.top6.motifs.stdThresh.tmo')
# Weights __MUST__ be in same order as motifs are loaded.
weights = [1,
2,
3,
4,
-1,
5]
assert len(motifs) == len(weights), \
'Error: len(motifs) MUST equal len(weights).'
motifs = zip(*[motifs, weights])
motifs.sort(key=lambda x: x[1])
motifs = zip(*motifs)
aligned = md.alignSimilarMotifs(motifs[0])
combined = sum(aligned,zip(*motifs[1]))
None
for h,s in zip(fg_fasta_headers,fg_fasta) :
h = h.replace('>'+get_org_settings(org)['genome_dir']+'/','')
h = h.replace('.nib','')
if len(s) > 150 :
fg_fasta_dict[h] = s
# now sample the background sequences
sys.stderr.write('Sampling bg sequences (len(fg_fasta)==%d)\n'%(len(fg_fasta_dict)))
#bg_fasta_dict = rejection_sample_bg(fg_fasta_dict,org,bg_match_epsilon=1e-3,verbose=True)
bg_fasta_dict = {}
bg_fasta = bg_fasta_dict.values()
"""
# load the motifs
sys.stderr.write('Movin right along\n')
motifs = load(motif_fn)
if opts.motif_ind != 'all' :
motif_indices = [int(i) for i in opts.motif_ind.split(',') if len(i) != 0]
motifs = [motifs[i] for i in motif_indices]
else :
motif_indices = xrange(len(motifs))
# use all cores w/ a Pool
#pool = Pool(processes=opts.n_procs)
# go through each motif
job_params = []
res = []
#for i,m in zip(motif_indices,motifs) :
# job_params.append((i,m,peak_pvals,fg_fasta,bg_fasta,opts.dir))