wpath = "dumps/wurcs"
gpath = "dumps/glycoct"
wlist = os.listdir(wpath)
glist = os.listdir(gpath)
alllist = list(set(wlist+glist))
print "Total glycan number %s" % len(alllist)
glycanobj = {}
for filename in alllist:
acc = filename.rstrip(".txt")
try:
gseq = open(os.path.join(gpath, filename)).read().strip()
obj = glycoct_parser.toGlycan(gseq)
except:
try:
wseq = open(os.path.join(wpath, filename)).read().strip()
obj = wurcs_parser.toGlycan(wseq)
except:
continue
glycanobj[acc] = obj
rpath = "dumps/redend"
reducing_end = {}
for filename in alllist:
acc = filename.rstrip(".txt")
r = open(os.path.join(rpath, filename)).read().strip()
glycan_length = {}
AllMotifpageid = AllMotif.id
for m in w.itermotif():
acc = m.get("glytoucan")
if acc in glycans:
continue
try:
glycans[acc] = wp.toGlycan(str(m.get("wurcs")))
except:
try:
glycans[acc] = gp.toGlycan(m.get("glycoct"))
except:
continue
g = glycans[acc]
l = len(list(g.all_nodes()))
glycan_length[acc] = l
print "%s motifs are supported" % len(glycans)
supported_acc = list(glycans.keys())
topology_pool = []
for i in range(len(supported_acc)):
for j in range(i, len(supported_acc)):
linkCheck = GlycanLinkCompatibleEitherway()
monoCheck = MonosaccharideCompatibleOneway()
rootMonoCheck = MonosaccharideCompatibleOneway()
if __name__ == "__main__":
seq1 = """RES
1b:x-dgal-HEX-1:5
2b:a-dgal-HEX-1:5
LIN
1:1o(3+1)2d"""
seq2 = """RES
1b:x-dglc-HEX-1:5
2s:n-acetyl
3b:b-dgal-HEX-1:5
4b:a-dgal-HEX-1:5
LIN
1:1d(2+1)2n
2:1o(4+1)3d
3:3o(4+1)4d"""
wurcsp = WURCS20Format()
glycoctp = GlycoCTFormat()
g1 = glycoctp.toGlycan(seq1)
g2 = glycoctp.toGlycan(seq2)
mstsa = MotifSearchTopologicalSameAs()
print mstsa.get(g1, g2)
strict_nred_matcher = pygly.alignment.NonReducingEndMotifStrict(
connected_nodes_cache=nodes_cache)
motif_gobjs = {}
for m in w.itermotif():
acc = m.get("glytoucan")
if acc in motif_gobjs:
continue
try:
motif_gobjs[acc] = wp.toGlycan(str(m.get("wurcs")))
except:
try:
motif_gobjs[acc] = gp.toGlycan(m.get("glycoct"))
except:
continue
archived = set()
gco = GlyCosmosNoCache()
for acc in gco.archived():
acc = acc["accession"]
archived.add(acc)
def secondtostr(i):
i = int(i)
h = i / 3600
m = (i - h * 3600) / 60
def substructure_search_init(shared_resources, structure_list_file_path, PPID):
print >> sys.stderr, "Computing Processor%s is starting" % PPID
task_queue, result_queue = shared_resources
gp = GlycoCTFormat()
wp = WURCS20Format()
motif_match_connected_nodes_cache = pygly.alignment.ConnectedNodesCache()
mm1 = pygly.alignment.GlyTouCanMotif(
connected_nodes_cache=motif_match_connected_nodes_cache)
# mm2 = pygly.alignment.MotifAllowOptionalSub(connected_nodes_cache=motif_match_connected_nodes_cache)
glycans = {}
for line in open(structure_list_file_path):
acc, s = line.strip().split()
glycans[acc] = wp.toGlycan(s)
print >> sys.stderr, "Processor-%s: finishes loading %s glycans" % (
PPID, len(glycans))
while True:
task_detail = task_queue.get(block=True)
print >> sys.stderr, "Processor-%s: Job %s received." % (
PPID, task_detail["id"])
seq = task_detail["seq"]
jobid = task_detail["id"]
#loose_root_match = task_detail["loose_root_match"]
#additional_subst = task_detail["additional_subst"]
motif_match_position = task_detail["motif_match_position"]
motif_matcher = mm1
"""
if loose_root_match:
motif_matcher = mm3
"""
#fullstructure = False
rootOnly = False
anywhereExceptRoot = False
if motif_match_position == "anywhere":
pass
elif motif_match_position == "reo":
rootOnly = True
else:
pass
"""
elif motif_match_position == "notre":
anywhereExceptRoot = True
elif motif_match_position == "fullstructure":
rootOnly = True
fullstructure = True
"""
matches = []
error = []
calculation_start_time = time.time()
try:
if "RES" in seq:
motif = gp.toGlycan(seq)
elif "WURCS" in seq:
motif = wp.toGlycan(seq)
else:
raise RuntimeError
except:
error.append("Unable to parse")
if len(error) == 0:
motif_node_num = len(list(motif.all_nodes()))
if motif_node_num > max_motif_size:
error.append("Motif is too big")
# TODO time out mechanism to avoid running for too long
for acc, glycan in glycans.items():
if len(error) != 0:
for e in error:
print >> sys.stderr, "Processor-%s: Issues (%s) is found with task %s" % (
PPID, e, task_detail["id"])
break
#if fullstructure:
# if motif_node_num != len(list(glycan.all_nodes())):
# continue
if motif_matcher.leq(motif,
glycan,
rootOnly=rootOnly,
anywhereExceptRoot=anywhereExceptRoot):
matches.append(acc)
calculation_end_time = time.time()
calculation_time_cost = calculation_end_time - calculation_start_time
res = {
"id": jobid,
"start time": calculation_start_time,
"end time": calculation_end_time,
"alignment calculation time": calculation_time_cost,
"matches": matches,
"error": error
}
print >> sys.stderr, "Processor-%s: Job %s finished within %ss" % (
PPID, task_detail["id"], calculation_time_cost)
result_queue.put(res)