def masterd4nonSp(prefix):
F = fasta(prefix + ".fasta")
ax1, ax2, ax3, nvec, allax = parse_eigenvectors(prefix)
pdbs, ind = get_gmfile(prefix)
equi = equi2dict(prefix)
ld = list2dict()
groupset = set(ld.values())
dgroup = {}
if pdbs and nvec == ind:
for e in range(len(pdbs)):
group = ld[equi[pdbs[e]]]
if not group in dgroup:
dgroup[group] = []
if dpc:
dgroup[group].append((F.chains[pdbs[e][:4]], allax[e][X - 1], allax[e][Y - 1], ax3[e]))
else:
dgroup[group].append((F.chains[pdbs[e][:4]], ax1[e], ax2[e], ax3[e]))
return dgroup, groupset
else:
print "Something is wrong with the files. Please check them and try again"
sys.exit()
def masterd4nonSp(prefix):
F = fasta(prefix + '.fasta')
ax1, ax2, ax3, nvec, allax = parse_eigenvectors(prefix)
pdbs, ind = get_gmfile(prefix)
equi = equi2dict(prefix)
ld = list2dict()
groupset = set(ld.values())
dgroup = {}
if pdbs and nvec == ind:
for e in range(len(pdbs)):
group = ld[equi[pdbs[e]]]
if not group in dgroup:
dgroup[group] = []
if dpc:
dgroup[group].append((F.chains[pdbs[e][:4]], allax[e][X - 1],
allax[e][Y - 1], ax3[e]))
else:
dgroup[group].append(
(F.chains[pdbs[e][:4]], ax1[e], ax2[e], ax3[e]))
return dgroup, groupset
else:
print 'Something is wrong with the files. Please check them and try again'
sys.exit()
def number_sps(prefix, db, level, remote):
"""
read a fasta and gm file file, and return a dictiorary the taxonomic level
(as in domain, class, etc...) as keys and the sp as values.
Also will return a set with the different groups.
"""
print "You choose taxonomic level %d:" % (level)
print "\t1 normally is superkingdom or domain"
print "\t>1 depends on the lineage reported in NCBI"
if not remote:
print "\nGoing locally:"
# import dictionary
P = Popen("echo $BLASTDB", shell=True, stderr=PIPE, stdout=PIPE)
PATH = P.communicate()[0].strip()
print "\tImporting dictionary with species and lineage from %s." % (PATH)
print "\tIf the path above is blank, set the $BLASTDB environment." " Make sure that the TaxDB.bin is in such path"
current = time.time()
d = pickle.load(open("%s/TaxDB.bin" % (PATH)))
print "\tTime elapsed to load the dictionary: %f" % (time.time() - current)
dgroup = {}
temp = []
# spslist=[]
F = fasta(prefix + ".fasta")
ax1, ax2, ax3, nvec, allax = parse_eigenvectors(prefix)
# if len(F.n) != nvec:
# print 'Some individuals are missing.'
print "Looking for the final gm file."
pdbs, ind = get_gmfile(prefix)
if pdbs and nvec == ind:
count = 1
for e in range(len(pdbs)):
if remote:
# print 'Sit thight, this will take a while, since it will go remotely to the NCBI taxnonomy database.'\
# ' Remember to run this in a weekend to avoid be blacklisted, or just because is going to take very'\
# ' long.'
while 1:
try:
group = entrez_query(db, F.chains[pdbs[e][:-1]], level)
break
except:
time.sleep(10)
else:
group = get_sp_locally(pdbs[e], d, level)
group = sentence_case(group)
if not group in dgroup:
dgroup[group] = []
if dpc:
dgroup[group].append((F.chains[pdbs[e][:4]], allax[e][X - 1], allax[e][Y - 1], ax3[e]))
else:
dgroup[group].append((F.chains[pdbs[e][:4]], ax1[e], ax2[e], ax3[e]))
temp.append(group)
# dgroup[group].sort()
# spslist.append(F.chains[e])
count += 1
s = set(temp)
return dgroup, s
else:
print "Something is wrong with the files. Please check them and try again"
sys.exit()
def plot(prefix, masterd, fontsize, symlog, threeD):
""" will plot two axes of the PCoA, colored by taxonomic level """
fil = open(prefix + "_PCoA_axes%d_%d.equivalences" % (X, Y), "w")
F = fasta(prefix + ".fasta")
markers = [
"k.",
"b+",
"g*",
"r.",
"c+",
"m*",
"y.",
"k+",
"b*",
"g.",
"r+",
"c*",
"m.",
"y+",
"k*",
"b.",
"g+",
"r*",
"c.",
"m+",
"y*",
]
count = 0
c = 0
fig = plt.figure() # figsize=(6.83 , 9.19), dpi=300)
ax = fig.add_subplot(111)
if threeD:
fig3D = plt.figure()
ax3D = fig3D.gca(projection="3d")
ax.spines["top"].set_color("none")
ax.xaxis.tick_bottom()
ax.spines["right"].set_color("none")
ax.yaxis.tick_left()
for k in masterd:
x = masterd[k][1]
y = masterd[k][2]
z = masterd[k][3]
ax.plot(x, y, markers[c], label=k)
if threeD:
ax3D.plot(x, y, z, markers[c], label=k)
c += 1
f = 0
for e in range(len(masterd[k][1])):
ax.annotate(count, (x[f] + 0.1, y[f] + 0.1), fontsize=fontsize)
if threeD:
ax3D.text(x[f] + 0.1, y[f] + 0.1, z[f] + 0.1, str(count), fontsize=fontsize)
fil.write(masterd[k][0][e] + "\t" + find_key(F.chains, masterd[k][0][e]) + "\t" + str(count) + "\n")
count += 1
f += 1
if symlog:
ax.set_xscale("symlog")
ax.set_yscale("symlog")
ax.set_xlabel("Axis %d (symmetrical log)" % (X), fontsize=fontsize)
ax.set_ylabel("Axis %d (symmetrical log)" % (Y), fontsize=fontsize)
else:
ax.set_xlabel("Axis %d" % (X), fontsize=fontsize)
ax.set_ylabel("Axis %d" % (Y), fontsize=fontsize)
ax.legend(loc=0, fancybox=True, shadow=True) # bbox_to_anchor=(0.5, 1.1), ncol=4,
fig.tight_layout()
if threeD:
ax3D.set_xlabel("Axis 1", fontsize=fontsize)
ax3D.set_ylabel("Axis 2", fontsize=fontsize)
ax3D.set_zlabel("Axis 3", fontsize=fontsize)
ax3D.view_init(30, 45)
fig3D.tight_layout()
ax3D.legend(loc=0, fancybox=True, shadow=True) # bbox_to_anchor=(0.5, -0.075),ncol=4,
plt.show()
fig.savefig(prefix + "_Axis%d_%dPCoA.png" % (X, Y), dpi=300)
if threeD:
fig3D.savefig(prefix + "_3AxesPCoA.png", dpi=300)
fil.close()
def number_sps(prefix, db, level, remote):
'''
read a fasta and gm file file, and return a dictiorary the taxonomic level
(as in domain, class, etc...) as keys and the sp as values.
Also will return a set with the different groups.
'''
print 'You choose taxonomic level %d:' % (level)
print '\t1 normally is superkingdom or domain'
print '\t>1 depends on the lineage reported in NCBI'
if not remote:
print "\nGoing locally:"
# import dictionary
P = Popen('echo $BLASTDB', shell=True, stderr=PIPE, stdout=PIPE)
PATH = P.communicate()[0].strip()
print '\tImporting dictionary with species and lineage from %s.' % (
PATH)
print '\tIf the path above is blank, set the $BLASTDB environment.'\
' Make sure that the TaxDB.bin is in such path'
current = time.time()
d = pickle.load(open('%s/TaxDB.bin' % (PATH)))
print '\tTime elapsed to load the dictionary: %f' % (time.time() -
current)
dgroup = {}
temp = []
#spslist=[]
F = fasta(prefix + '.fasta')
ax1, ax2, ax3, nvec, allax = parse_eigenvectors(prefix)
#if len(F.n) != nvec:
#print 'Some individuals are missing.'
print 'Looking for the final gm file.'
pdbs, ind = get_gmfile(prefix)
if pdbs and nvec == ind:
count = 1
for e in range(len(pdbs)):
if remote:
#print 'Sit thight, this will take a while, since it will go remotely to the NCBI taxnonomy database.'\
# ' Remember to run this in a weekend to avoid be blacklisted, or just because is going to take very'\
# ' long.'
while 1:
try:
group = entrez_query(db, F.chains[pdbs[e][:-1]], level)
break
except:
time.sleep(10)
else:
group = get_sp_locally(pdbs[e], d, level)
group = sentence_case(group)
if not group in dgroup:
dgroup[group] = []
if dpc:
dgroup[group].append((F.chains[pdbs[e][:4]], allax[e][X - 1],
allax[e][Y - 1], ax3[e]))
else:
dgroup[group].append(
(F.chains[pdbs[e][:4]], ax1[e], ax2[e], ax3[e]))
temp.append(group)
#dgroup[group].sort()
#spslist.append(F.chains[e])
count += 1
s = set(temp)
return dgroup, s
else:
print 'Something is wrong with the files. Please check them and try again'
sys.exit()
def plot(prefix, masterd, fontsize, symlog, threeD):
''' will plot two axes of the PCoA, colored by taxonomic level '''
fil = open(prefix + '_PCoA_axes%d_%d.equivalences' % (X, Y), 'w')
F = fasta(prefix + '.fasta')
markers = [
'k.', 'b+', 'g*', 'r.', 'c+', 'm*', 'y.', 'k+', 'b*', 'g.', 'r+', 'c*',
'm.', 'y+', 'k*', 'b.', 'g+', 'r*', 'c.', 'm+', 'y*'
]
count = 0
c = 0
fig = plt.figure() #figsize=(6.83 , 9.19), dpi=300)
ax = fig.add_subplot(111)
if threeD:
fig3D = plt.figure()
ax3D = fig3D.gca(projection='3d')
ax.spines['top'].set_color('none')
ax.xaxis.tick_bottom()
ax.spines['right'].set_color('none')
ax.yaxis.tick_left()
for k in masterd:
x = masterd[k][1]
y = masterd[k][2]
z = masterd[k][3]
ax.plot(x, y, markers[c], label=k)
if threeD:
ax3D.plot(x, y, z, markers[c], label=k)
c += 1
f = 0
for e in range(len(masterd[k][1])):
ax.annotate(count, (x[f] + 0.1, y[f] + 0.1), fontsize=fontsize)
if threeD:
ax3D.text(x[f] + 0.1,
y[f] + 0.1,
z[f] + 0.1,
str(count),
fontsize=fontsize)
fil.write(masterd[k][0][e] + '\t' +
find_key(F.chains, masterd[k][0][e]) + '\t' +
str(count) + '\n')
count += 1
f += 1
if symlog:
ax.set_xscale("symlog")
ax.set_yscale("symlog")
ax.set_xlabel('Axis %d (symmetrical log)' % (X), fontsize=fontsize)
ax.set_ylabel('Axis %d (symmetrical log)' % (Y), fontsize=fontsize)
else:
ax.set_xlabel('Axis %d' % (X), fontsize=fontsize)
ax.set_ylabel('Axis %d' % (Y), fontsize=fontsize)
ax.legend(
loc=0, #bbox_to_anchor=(0.5, 1.1), ncol=4,
fancybox=True,
shadow=True)
fig.tight_layout()
if threeD:
ax3D.set_xlabel('Axis 1', fontsize=fontsize)
ax3D.set_ylabel('Axis 2', fontsize=fontsize)
ax3D.set_zlabel('Axis 3', fontsize=fontsize)
ax3D.view_init(30, 45)
fig3D.tight_layout()
ax3D.legend(
loc=0, #bbox_to_anchor=(0.5, -0.075),ncol=4,
fancybox=True,
shadow=True)
plt.show()
fig.savefig(prefix + '_Axis%d_%dPCoA.png' % (X, Y), dpi=300)
if threeD:
fig3D.savefig(prefix + '_3AxesPCoA.png', dpi=300)
fil.close()
