def gen_sam_pair(fpath1, fpath2, flip_1, flip_2):
''' pulls info from paired sam file '''
fin1 = open(fpath1)
fin2 = open(fpath2)
for line1 in fin1:
line2 = fin2.readline()
# sanity.
if line1[0] == "@":
continue
# tokenize.
tok1 = line1.strip().split()
tok2 = line2.strip().split()
if len(tok1) < 2 or len(tok2) < 2:
print tok1, tok2
# simplify.
if tok1[1] == '0':
orien1 = 0
else:
orien1 = 1
if tok2[1] == '0':
orien2 = 0
else:
orien2 = 1
if flip_1 == True:
orien1 = 1 - orien1
if flip_2 == True:
orien2 = 1 - orien2
ctg1 = tok1[2]
ctg2 = tok2[2]
left1 = int(tok1[3])
left2 = int(tok2[3])
right1 = left1 + len(tok1[9])
right2 = left2 + len(tok2[9])
print ctg1<ctg2, orien1, orien2
state = misc.determine_state(ctg1, ctg2, orien1, orien2)
# yield the info.
yield ctg1, left1, right1, ctg2, left2, right2, state, tok1, tok2
fin1.close()
fin2.close()
def agp_graph_undirected(fpath):
''' returns agp graph '''
# load agp array.
agp_edges = load_agp(fpath)
# make digraph.
G = nx.Graph()
# add nodes.
for i in range(agp_edges.size):
# skip contigs themselves.
if agp_edges[i]['comp_type'] != 'W': continue
# add node info.
name = agp_edges[i]['comp_name']
width = agp_edges[i]['comp_stop']
orien = agp_edges[i]['comp_orien']
G.add_node(agp_edges[i]['comp_name'], {'width':width, 'orien':orien})
# add edges.
for i in range(agp_edges.size):
# skip contigs themselves.
if agp_edges[i]['comp_type'] != 'N': continue
if i == agp_edges.shape[0] - 1: continue
# add sorted edges.
ctg1 = agp_edges[i-1]['comp_name']
ctg2 = agp_edges[i+1]['comp_name']
o1 = G.node[ctg1]['orien']
o2 = G.node[ctg2]['orien']
gap = agp_edges[i]['comp_stop']
state = misc.determine_state(ctg1, ctg2, o1, o2)
G.add_edge(ctg1, ctg2, {'gap':gap, 'state':state})
# done.
return G