def checknsample(prefix, groups,eq,gmfile):
linespg={}
lenght=[]
namesingm=[]
for k, va in groups.iteritems():
lenght.append(len(va))
linespg[k]=[]
for l in open(prefix+'.gm'):
for sn in va:
if sn == '':
continue
if eq[sn] in l:
linespg[k].append(l)
minimum=min(lenght)
#minimum=40 #workaround to get contacts... uncomment previous line to restore
move(prefix+'.gm',prefix+'_beforesampling.gm')
landm = land2dict(prefix+'.landmarks')
move(prefix+'.landmarks',prefix+'_beforesampling.landmarks')
fout=open(prefix+'.gm','w')
outf=open(prefix+'.landmarks','w')
for ke, val in linespg.iteritems():
gout = open(ke+'.gm','w')
S = sample(val,minimum)
for i in S:
fout.write(i)
gout.write(i)
namesingm.append(find_key(eq,i[1:i.find(':')]))
for key, value in landm.iteritems():
if key in i:
outf.write('>'+key+'\n'+value+'\n')
gout.close()
return namesingm
def checknsample(prefix, groups, eq, gmfile):
linespg = {}
lenght = []
namesingm = []
for k, va in groups.iteritems():
lenght.append(len(va))
linespg[k] = []
for l in open(prefix + '.gm'):
for sn in va:
if sn == '':
continue
if eq[sn] in l:
linespg[k].append(l)
minimum = min(lenght)
#minimum=40 #workaround to get contacts... uncomment previous line to restore
move(prefix + '.gm', prefix + '_beforesampling.gm')
landm = land2dict(prefix + '.landmarks')
move(prefix + '.landmarks', prefix + '_beforesampling.landmarks')
fout = open(prefix + '.gm', 'w')
outf = open(prefix + '.landmarks', 'w')
for ke, val in linespg.iteritems():
gout = open(ke + '.gm', 'w')
S = sample(val, minimum)
for i in S:
fout.write(i)
gout.write(i)
namesingm.append(find_key(eq, i[1:i.find(':')]))
for key, value in landm.iteritems():
if key in i:
outf.write('>' + key + '\n' + value + '\n')
gout.close()
return namesingm
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 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()
