def main(ddir):
tree = common.read_tree_adj_list(ddir)
scores, all_ecs = common.read_scores(ddir, augmented_format=True)
ecscores = {}
max_blast = 0
for species in scores:
for ec in scores[species]:
if ec not in ecscores:
ecscores[ec] = {}
ecscores[ec][species] = scores[species][ec]
blast = scores[species][ec].bscore
if blast > max_blast:
max_blast = blast
try:
os.mkdir("%s/%s" % (ddir, common.EC_VISUALS_DIR))
except:
pass
keys = ecscores.keys()
keys.sort()
for ec in keys:
#print ec
s = tree_to_dot(tree, ecscores[ec], max_blast, ec)
#print s
outdir = "%s/%s" % (ddir, common.EC_VISUALS_DIR)
outfn = "%s" % (ec)
plot(s, outdir, outfn)
def main(ddir):
tree = common.read_tree_adj_list(ddir)
scores, all_ecs = common.read_scores(ddir, augmented_format = True)
ecscores = {}
max_blast = 0
for species in scores:
for ec in scores[species]:
if ec not in ecscores:
ecscores[ec] = {}
ecscores[ec][species] = scores[species][ec]
blast = scores[species][ec].bscore
if blast > max_blast:
max_blast = blast
try:
os.mkdir("%s/%s" % (ddir, common.EC_VISUALS_DIR))
except:
pass
keys = ecscores.keys()
keys.sort()
for ec in keys:
#print ec
s = tree_to_dot(tree, ecscores[ec], max_blast, ec)
#print s
outdir = "%s/%s" % (ddir, common.EC_VISUALS_DIR)
outfn = "%s" % (ec)
plot(s, outdir, outfn)
def roc(ddir, mdir, remove_partial=FILTER_PARTIAL_ECS):
print "Reading models..."
models, model_ecs = common.read_models(mdir, remove_partial)
print "Reading reaction scores..."
scores, data_ecs = common.read_scores(ddir, True)
if remove_partial:
data_ecs = filter_partial_ecs(data_ecs)
all_ecs = model_ecs.union(data_ecs)
print "Computing ROC curves..."
compute_roc_curves(models, scores, all_ecs)
def roc(ddir, mdir, remove_partial = FILTER_PARTIAL_ECS):
print "Reading models..."
models, model_ecs = common.read_models(mdir, remove_partial)
print "Reading reaction scores..."
scores, data_ecs = common.read_scores(ddir, True)
if remove_partial:
data_ecs = filter_partial_ecs(data_ecs)
all_ecs = model_ecs.union(data_ecs)
print "Computing ROC curves..."
compute_roc_curves(models, scores, all_ecs)
def roc_one(ddir, modelfn, species, outdir, remove_partial=FILTER_PARTIAL_ECS):
roc_dir = "%s/%s" % (ddir, outdir)
try:
os.mkdir(roc_dir)
except:
pass
model = common.read_model(open(modelfn), remove_partial)
scores, data_ecs = common.read_scores(ddir, True)
if remove_partial:
data_ecs = filter_partial_ecs(data_ecs)
all_ecs = model.union(data_ecs)
models = {species: model}
of = open("%s/%s.full" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"pscore", 1, remove_partial)
of.close()
of = open("%s/%s.naive" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"npscore", 1, remove_partial)
of.close()
of = open("%s/%s.blast" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"bscore", None, remove_partial)
of.close()
of = open("%s/%s.gtg" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"gscore", 1, remove_partial)
of.close()
of = open("%s/%s.blasttree" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"btscore", 1, remove_partial)
of.close()
of = open("%s/%s.gtgtree" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"gtscore", 1, remove_partial)
of.close()
plot_single(ddir, species, outdir, "pdf")
plot_single(ddir, species, outdir, "png")
subprocess.call("rm *.Rout", shell=True)
def roc_one(ddir, modelfn, species, outdir, remove_partial = FILTER_PARTIAL_ECS):
roc_dir = "%s/%s" % (ddir, outdir)
try:
os.mkdir(roc_dir)
except:
pass
model = common.read_model(open(modelfn), remove_partial)
scores, data_ecs = common.read_scores(ddir, True)
if remove_partial:
data_ecs = filter_partial_ecs(data_ecs)
all_ecs = model.union(data_ecs)
models = {species : model}
of = open("%s/%s.full" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"pscore", 1, remove_partial)
of.close()
of = open("%s/%s.naive" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"npscore", 1, remove_partial)
of.close()
of = open("%s/%s.blast" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"bscore", None, remove_partial)
of.close()
of = open("%s/%s.gtg" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"gscore", 1, remove_partial)
of.close()
of = open("%s/%s.blasttree" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"btscore", 1, remove_partial)
of.close()
of = open("%s/%s.gtgtree" % (roc_dir, species), "w")
roc_curves = compute_roc_curve(models, scores, all_ecs, of, species,
"gtscore", 1, remove_partial)
of.close()
plot_single(ddir, species, outdir, "pdf")
plot_single(ddir, species, outdir, "png")
subprocess.call("rm *.Rout", shell = True)
def compute_reaction_scores(ddir,
r2ec,
ec2r,
target_species,
remove_partial=False):
scores, all_ecs = common.read_scores(ddir, True)
try:
os.mkdir("%s/%s" % (ddir, common.REACTION_SCORE_DIR))
except:
pass
print "Computing reaction scores..."
for species in scores:
if target_species != None and species not in target_species:
continue
print " %s" % (species)
sfn = "%s/%s/%s" % (ddir, common.REACTION_SCORE_DIR, species)
o = open(sfn, "w")
rscores = {}
rec = {}
for ec in scores[species]:
if remove_partial and "-" in ec:
print "reco-dir.py: Discarding reaction score for partial EC number:", ec
continue
score = scores[species][ec]
if ec in ec2r:
for r in ec2r[ec]:
if r not in rec:
rec[r] = set()
if r not in rscores or score.pscore > rscores[r].pscore:
rscores[r] = score
rec[r].add(ec)
else:
o.write("?\t%s\t%s\n" % (ec, score))
rs = rscores.keys()
rs.sort()
for r in rs:
o.write("%s\t%s\t%s\n" % (r, ",".join(rec[r]), rscores[r]))
o.close()
return scores.keys()
def plot_reaction_scores(ddir):
scores, all_ecs = common.read_scores(ddir, augmented_format = True)
keys = scores.keys()
keys.sort()
try:
os.mkdir("%s/%s" % (ddir, common.REACTION_SCORE_PLOT_DIR))
except:
pass
for species in keys:
sys.stdout.write("%s " % (species))
plotfn = "%s/%s/%s" %(ddir, common.REACTION_SCORE_PLOT_DIR, species)
tf = tempfile.NamedTemporaryFile()
o = open(tf.name, "w")
for ec in scores[species]:
rscore = scores[species][ec]
o.write("%s\n" % (rscore))
o.flush()
splot(tf.name, plotfn, "pdf")
splot(tf.name, plotfn, "png")
o.close()
def plot_reaction_scores(ddir):
scores, all_ecs = common.read_scores(ddir, augmented_format=True)
keys = scores.keys()
keys.sort()
try:
os.mkdir("%s/%s" % (ddir, common.REACTION_SCORE_PLOT_DIR))
except:
pass
for species in keys:
sys.stdout.write("%s " % (species))
plotfn = "%s/%s/%s" % (ddir, common.REACTION_SCORE_PLOT_DIR, species)
tf = tempfile.NamedTemporaryFile()
o = open(tf.name, "w")
for ec in scores[species]:
rscore = scores[species][ec]
o.write("%s\n" % (rscore))
o.flush()
splot(tf.name, plotfn, "pdf")
splot(tf.name, plotfn, "png")
o.close()
def compute_reaction_scores(ddir, r2ec, ec2r, target_species, remove_partial = False):
scores, all_ecs = common.read_scores(ddir, True)
try:
os.mkdir("%s/%s" % (ddir, common.REACTION_SCORE_DIR))
except:
pass
print "Computing reaction scores..."
for species in scores:
if target_species != None and species not in target_species:
continue
print " %s" % (species)
sfn = "%s/%s/%s" % (ddir, common.REACTION_SCORE_DIR, species)
o = open(sfn, "w")
rscores = {}
rec = {}
for ec in scores[species]:
if remove_partial and "-" in ec:
print "reco-dir.py: Discarding reaction score for partial EC number:", ec
continue
score = scores[species][ec]
if ec in ec2r:
for r in ec2r[ec]:
if r not in rec:
rec[r] = set()
if r not in rscores or score.pscore > rscores[r].pscore:
rscores[r] = score
rec[r].add(ec)
else:
o.write("?\t%s\t%s\n" % (ec, score))
rs = rscores.keys()
rs.sort()
for r in rs:
o.write("%s\t%s\t%s\n" % (r, ",".join(rec[r]), rscores[r]))
o.close()
return scores.keys()
#!/usr/bin/env python
import sys, os, datetime
import common
ddir = sys.argv[1] # project dir
scoredir = sys.argv[2] # dir with .ecscores
o = open("%s/%s" % (ddir, common.FULL_REACTION_SCORE_FILE), "w")
rscores, ecs = common.read_scores(ddir)
org_abbr, org_long = common.read_organisms(ddir)
fns = os.listdir(scoredir)
scores = {}
for fn in fns:
if fn.endswith(".ecscores"):
org = org_abbr[fn[:-len(".ecscores")]]
scores[org] = {}
f = open("%s/%s" % (scoredir, fn))
for s in f:
if s.startswith("#"):
continue
ec, bs, bs1, bs2, gs, gs1, gs2 = s.strip().split("\t")
scores[org][ec] = (bs, bs1, bs2, gs, gs1, gs2)
ecs.add(ec)
keys = set(rscores.keys())
#return set() keys with adding set(scores.keys())
keys.update(set(scores.keys()))
keys = list(keys)
#!/usr/bin/env python
import sys, os, datetime
import common
ddir = sys.argv[1] # project dir
scoredir = sys.argv[2] # dir with .ecscores
o = open("%s/%s" % (ddir, common.FULL_REACTION_SCORE_FILE), "w")
rscores, ecs = common.read_scores(ddir)
org_abbr, org_long = common.read_organisms(ddir)
fns = os.listdir(scoredir)
scores = {}
for fn in fns:
if fn.endswith(".ecscores"):
org = org_abbr[fn[:-len(".ecscores")]]
scores[org] = {}
f = open("%s/%s" % (scoredir, fn))
for s in f:
if s.startswith("#"):
continue
ec, bs, bs1, bs2, gs, gs1, gs2 = s.strip().split("\t")
scores[org][ec] = (bs, bs1, bs2, gs, gs1, gs2)
ecs.add(ec)
keys = set(rscores.keys())
#return set() keys with adding set(scores.keys())
keys.update(set(scores.keys()))
keys = list(keys)