def save_tables_show_graph(self, desk):
if desk.each_all == 'each':
bias_corr_list = [desk.withCorrection]
else:
bias_corr_list = [False, True]
print("-----------------------------------------------------------")
for bias_corr in bias_corr_list:
iLoop = 0
speciesParams = []
listMI_Anova = []
for mat in self.list_species_name_hmi:
iLoop += 1
filename, name, species, hmi, sufix = mat
mat = desk.dicParams[species]
if desk.minmax == 'mincut':
numOfSeqs = mat[2]
L = mat[5]
else:
numOfSeqs = mat[3]
L = mat[6]
print("%i) %s - %s, %s" %
(iLoop, name, desk.minmax,
"bias corr." if bias_corr else 'no corr.'))
if desk.organism == 'Drosophila':
sp = self.dr.mnemonic(species.replace('Drosophila ', ''))
else:
sp = species
desk.withCorrection = bias_corr
if desk.withCorrection:
str_correction = ' (bias corr.)'
filename_correction = '_bias_corr'
else:
str_correction = ''
filename_correction = ''
if desk.frame == 0:
sFrame = ''
else:
sFrame = ', frame=' + str(desk.frame)
std_rand = "" if (desk.label_random
== "") else " " + desk.label_random
title = 'HMI: %s %s, %s %s\n%s%s%s; %i seqs; len=%i%s; #letter=%i'%\
(desk.organism, species, desk.seqType, desk.gene_title, desk.minmax, str_correction, std_rand, numOfSeqs, L, sFrame, desk.numOfLetters)
''' HMI mean and sdv for each k from n sequences '''
if not desk.withCorrection:
arrayMIthis = np.array(hmi.arrayMI)
arraySE = np.array(hmi.arraySE)
else:
arrayMIthis = np.array(hmi.arrayMIcorr)
arraySE = np.array(hmi.arraySEcorr)
if desk.mnat:
arrayMIthis *= 1000
arraySE *= 1000
roundVal = 1
unit = 'mnat'
else:
roundVal = 4
unit = 'nat'
if desk.norm:
arrayMIthis /= desk.numOfLetters
arraySE /= desk.numOfLetters
# desk.showmsg_obs(' >>> species %s with %i sequences'%(species, numOfSeqs))
''' first time build abcissa
if choose == 1 and desk.frame == 1
or desk.frame == 0
'''
xSeq = []
if desk.frame < 2:
for k in range(len(arrayMIthis)):
xSeq.append(k + 3)
else:
''' frame1: k = {3,6,9, ... 3n} '''
for k in range(len(arrayMIthis)):
xSeq.append((k + 1) * 3)
arrayVal = []
ySeqSEFinal = []
''' if only one curve '''
if desk.frame < 2:
arrayVal = arrayMIthis
ySeqSEFinal = arraySE
else:
''' if choose > 1 there are 3 others curves, and k jumps 3 x 3 '''
for m in range(len(arrayMIthis)):
arrayVal.append(arrayMIthis[m])
arrayVal.append(arrayMIthis[m])
arrayVal.append(arrayMIthis[m])
ySeqSEFinal.append(arraySE[m])
ySeqSEFinal.append(arraySE[m])
ySeqSEFinal.append(arraySE[m])
self.meanY = np.round(np.mean(arrayVal), roundVal)
self.medianY = np.round(np.median(arrayVal), roundVal)
self.stdY = np.round(np.std(arrayVal), roundVal)
self.maxY = np.round(np.max(arrayVal), roundVal)
self.minY = np.round(np.min(arrayVal), roundVal)
speciesParams.append([
species, sp, numOfSeqs, self.meanY, self.medianY,
self.stdY, self.maxY, self.minY
])
listMI_Anova.append(arrayVal)
if desk.showGraph or desk.saveGraph:
if desk.frame < 2:
gHist = graphPac.Histogram_FreqDistribution(
desk, title)
gHist.meanY = self.meanY
gHist.medianY = self.medianY
gHist.stdY = self.stdY
gHist.maxY = self.maxY
gHist.minY = self.minY
gHist.plot_H_MI('V',
xSeq=xSeq,
ySeq=arrayVal,
ySE=arraySE,
showError=True)
gHist.plot_H_MI('H',
xSeq=xSeq,
ySeq=arrayVal,
ySE=ySeqSEFinal,
showError=True)
gHist.densityBar("H", seq=arrayVal)
else:
gHist.sameBar(
xSeq=xSeq,
arrayMIthis=arrayVal,
linestyleCode=self.arrLinestyleCode[desk.frame],
color=self.arrColor[desk.frame])
if (desk.frame == 0) or (desk.frame == 3):
pictureName = 'HMI_%s%s' % (sufix, filename_correction)
gHist.myPlot.print_graph(desk,
gHist.fig,
pictureName,
frame=desk.tk_root)
'''
del gHist
plt.cla()
plt.clf()
gc.collect()
'''
if desk.saveData:
sufix = ('%s_%s_%s_%s_frame%i_NOL%i_%iL_cutoff%i') %\
(desk.organism, desk.minmax, desk.seqType, desk.gene_title, desk.frame, desk.numOfLetters, desk.cutoffLength, desk.cutoffNumSeq)
std_rand = "" if (desk.label_random
== "") else desk.stri_random
filename = 'HMI_summary_%s%s%s.txt' % (
sufix, filename_correction, std_rand)
stri = hmi.ent.calc_anova(listMI_Anova, sufix)
hmi.ent.print_data_summary(desk,
speciesParams,
roundVal=roundVal,
filename=filename,
stri=stri)
self.failure = False
self.error_msg = 'Task ended. All right.'
return True
def save_tables_show_graph(self,cmp_file=""):
bias_corr_list = [self.withCorrection]
print("-----------------------------------------------------------")
for bias_corr in bias_corr_list:
iLoop = 0
speciesParams = []
listMI_Anova = []
for mat in self.list_species_name_hmi:
iLoop += 1
filename, name, species, hmi, sufix = mat
numOfSeqs = self.sampleElems
L = self.LH0
print("%i) %s - %s, %s"%(iLoop, name, self.minmax, "bias corr." if bias_corr else 'no corr.' ))
sp = self.which_sp(self.organism, species)
self.withCorrection = bias_corr
if self.withCorrection:
str_correction = ' (bias corr.)'
filename_correction = '_bias_corr'
else:
str_correction = ''
filename_correction = ''
if self.frame == 0:
sFrame = ''
else:
sFrame = ', frame=' + str(self.frame)
title = 'HMI: %s %s, %s %s\n%s%s; %i seqs; len=%i%s; #letter=%i'%\
(self.organism, species, self.seqType, self.gene, self.minmax, str_correction, numOfSeqs, L, sFrame, self.numOfLetters)
''' HMI mean and sdv for each k from n sequences '''
if not self.withCorrection:
arrayMIthis = np.array(hmi.arrayMI)
arraySE = np.array(hmi.arraySE)
else:
arrayMIthis = np.array(hmi.arrayMIcorr)
arraySE = np.array(hmi.arraySEcorr)
if self.mnat:
arrayMIthis *= 1000
arraySE *= 1000
roundVal = 1
self.unit = 'mnat'
else:
roundVal = 4
self.unit = 'nat'
if self.norm:
arrayMIthis /= self.numOfLetters
arraySE /= self.numOfLetters
# self.showmsg_obs(' >>> species %s with %i sequences'%(species, numOfSeqs))
''' first time build abcissa
if choose == 1 and self.frame == 1
or self.frame == 0
'''
xSeq = []
if self.frame < 2:
for k in range( len(arrayMIthis)):
xSeq.append(k+3)
else:
''' frame1: k = {3,6,9, ... 3n} '''
for k in range( len(arrayMIthis)):
xSeq.append((k+1)*3)
arrayVal = []
ySeqSEFinal = []
''' if only one curve '''
if self.frame < 2:
arrayVal = arrayMIthis
ySeqSEFinal = arraySE
else:
''' if choose > 1 there are 3 others curves, and k jumps 3 x 3 '''
for m in range(len(arrayMIthis)):
arrayVal.append(arrayMIthis[m])
arrayVal.append(arrayMIthis[m])
arrayVal.append(arrayMIthis[m])
ySeqSEFinal.append(arraySE[m])
ySeqSEFinal.append(arraySE[m])
ySeqSEFinal.append(arraySE[m])
self.meanY = np.round(np.mean(arrayVal), roundVal)
self.medianY = np.round(np.median(arrayVal), roundVal)
self.stdY = np.round(np.std(arrayVal), roundVal)
self.maxY = np.round(np.max(arrayVal), roundVal)
self.minY = np.round(np.min(arrayVal), roundVal)
speciesParams.append([species,sp,numOfSeqs, self.meanY, self.medianY, self.stdY, self.maxY, self.minY])
listMI_Anova.append(arrayVal)
if self.showGraph or self.saveGraph:
if self.frame < 2:
gHist = graphPac.Histogram_FreqDistribution(self, title)
gHist.meanY = self.meanY
gHist.medianY = self.medianY
gHist.stdY = self.stdY
gHist.maxY = self.maxY
gHist.minY = self.minY
gHist.desk = self
gHist.plot_H_MI('V', xSeq=xSeq, ySeq=arrayVal, ySE=arraySE, showError=True, \
unit=self.unit, roundVal=roundVal)
gHist.plot_H_MI('H', xSeq=xSeq, ySeq=arrayVal, ySE=ySeqSEFinal, showError=True, \
unit=self.unit, roundVal=roundVal)
gHist.densityBar("H", seq=arrayVal, unit=self.unit, roundVal = roundVal)
else:
gHist.sameBar(xSeq=xSeq, arrayMIthis=arrayVal,linestyleCode=self.arrLinestyleCode[self.frame], color=self.arrColor[self.frame])
if (self.frame == 0) or (self.frame==3):
''' sample in filename '''
pictureName = 'HMI_%s%s%s'%(sufix, filename_correction, cmp_file)
gHist.myPlot.print_graph(self, gHist.fig, pictureName, frame=self.tk_root, stay=True)
'''
del gHist
plt.cla()
plt.clf()
gc.collect()
'''
if self.saveData :
sufix = ('%s_%s_%s_%s_frame%i_NOL%i_%iL_cutoff%i') %\
(self.organism, self.minmax, self.seqType, self.gene, self.frame, self.numOfLetters, self.cutoffLength, self.cutoffNumSeq)
''' summary with sample in filename '''
filename = 'HMI_summary_%s%s%s.txt' % (sufix, filename_correction, cmp_file)
stri = hmi.ent.calc_anova(listMI_Anova, sufix)
hmi.ent.print_data_summary(self, speciesParams, roundVal=roundVal, filename=filename, stri=stri)
self.failure = False
self.error_msg = 'Task ended. All right.'
''' in the end of save_tables_show_graph clean memory '''
del self.list_species_name_hmi
try:
gc.collect()
except:
pass
return True
def save_tables_show_graph(self, desk, which_module):
if desk.each_all == 'each':
list_modules = [which_module]
bias_corr_list = [desk.withCorrection]
else:
list_modules = ['Entropy', 'VMI']
bias_corr_list = [False, True]
print("-----------------------------------------------------------")
for which_module in list_modules:
for bias_corr in bias_corr_list:
iLoop = 0
listMI_Anova = []
speciesParams = []
for mat in self.list_species_name_vmi:
iLoop += 1
filename, name, species, vmi, sufix = mat
mat = desk.dicParams[species]
if desk.minmax == 'mincut':
numOfSeqs = mat[2]
L = mat[5]
else:
numOfSeqs = mat[3]
L = mat[6]
print("%i) %s - %s %s %s"%(iLoop, name, desk.minmax, "bias corr." if bias_corr else 'no corr.', which_module ))
if desk.organism == 'Drosophila':
sp = self.dr.mnemonic(species.replace('Drosophila ',''))
else:
sp = species
desk.withCorrection = bias_corr
if desk.withCorrection:
str_correction = ' (bias corr.)'
filename_correction = '_bias_corr'
else:
str_correction = ''
filename_correction = ''
std_rand = "" if (desk.label_random == "") else " " + desk.label_random
if which_module == 'Entropy':
title = 'Entropy Distribution %s %s, %s %s\n%s%s%s; %i seqs; len=%i; #letter=%i'%\
(desk.organism, species, desk.seqType, desk.gene_title, desk.minmax, str_correction, std_rand, numOfSeqs, L, desk.numOfLetters)
else:
title = 'VMI Heat Map %s %s, %s %s %s%s%s\n%i seqs; len=%i; #letter=%i'%\
(desk.organism, species, desk.seqType, desk.gene_title, desk.minmax, str_correction, std_rand, numOfSeqs, L, desk.numOfLetters)
if which_module == 'Entropy':
if not desk.withCorrection:
arrayVal = np.array(vmi.HShannonList)
arraySE = np.array(vmi.SeHShannonList)
else:
arrayVal = np.array(vmi.HShannonCorrList)
arraySE = np.array(vmi.SeHShannonCorrList)
else:
if not desk.withCorrection:
arrayVal = np.array(vmi.MIlist)
arraySE = np.array(vmi.SeMIList)
else:
arrayVal = np.array(vmi.MIcorrList)
arraySE = np.array(vmi.SeMICorrList)
''' normalization dividing by numOfLetters '''
if desk.norm:
arrayVal /= desk.numOfLetters
arraySE /= desk.numOfLetters
''' mili nats '''
if desk.mnat:
arrayVal *= 1000
arraySE *= 1000
roundVal = 2
else:
roundVal = 4
# print '--- params -----------------'
maxMI = 0
maxiPos = None
is_zero = True
for pos in range(len(arrayVal)):
if arrayVal[pos] > maxMI:
maxMI = arrayVal[pos]
SE = arraySE[pos]
i,j = vmi.ijList[pos]
maxiPos = [i,j,maxMI,SE]
is_zero = False
if is_zero:
stri = '### Species %s has MI = ZERO. Too conserved data sequences. Impossible to include in analysis.'%(species)
print(stri)
continue
xSeq = [x for x in range(len(arrayVal))]
self.meanY = np.round(np.mean(arrayVal), roundVal)
self.medianY = np.round(np.median(arrayVal), roundVal)
self.stdY = np.round(np.std(arrayVal), roundVal)
self.maxY = np.round(np.max(arrayVal), roundVal)
self.minY = np.round(np.min(arrayVal), roundVal)
speciesParams.append([species,sp,numOfSeqs, self.meanY, self.medianY, self.stdY, self.maxY, self.minY])
if desk.saveData:
listMI_Anova.append(arrayVal)
if desk.showGraph or desk.saveGraph:
if which_module == 'Entropy':
gHist = graphPac.Histogram_FreqDistribution(desk, title)
gHist.meanY = self.meanY
gHist.medianY = self.medianY
gHist.stdY = self.stdY
gHist.maxY = self.maxY
gHist.minY = self.minY
gHist.plot_H_MI('V', xSeq=xSeq, ySeq=arrayVal, ySE=arraySE, showError=True)
gHist.densityBar('V', seq=arrayVal)
pictureName = 'VHShannon_%s%s%s'%(sufix, filename_correction, desk.stri_random)
gHist.myPlot.print_graph(desk, gHist.fig, pictureName, frame=desk.tk_root)
else:
''' 3D dont has ceil '''
if desk.is3D:
limSup = self.maxY
else:
ceil = desk.heatmap_ceil_value
''' the same roof for all heatmaps: parametrize in future '''
if desk.heatmap_ceil:
if self.maxY <= ceil:
limSup = ceil
else:
limSup = self.maxY
else:
limSup = self.maxY
'''' updated 28/09/2015 '''
gMI = vmi.plotHeatMap(desk, desk.is3D, arrayVal, vmi.ijList, L, maxiPos, title, species=species, limSup=limSup, roundVal=roundVal, str_correction=str_correction)
if desk.is3D:
pictureName = 'HeatMap_3D_VMI_%s%s%s'%(sufix, filename_correction,desk.stri_random)
else:
gMI.densityHeatmapBar(arrayVal, limSup)
pictureName = 'HeatMap_2D_VMI_%s%s%s'%(sufix, filename_correction,desk.stri_random)
gMI.myPlot.print_graph(desk, gMI.fig, pictureName, frame=desk.tk_root)
'''
plt.cla()
plt.clf()
plt.close()
del gMI
gc.collect()
'''
# Flavio 02/06/2015
if which_module == 'Entropy':
which_symb = 'VHShannon'
else:
which_symb = 'VMI'
if desk.saveData:
sufix = ('%s_%s_%s_%s_NOL%i_%iL_cutoff%i') %\
(desk.organism, desk.minmax, desk.seqType, desk.gene_title, desk.numOfLetters, desk.cutoffLength, desk.cutoffNumSeq)
filename = '%s_summary_%s%s%s.txt' % (which_symb, sufix, filename_correction, desk.stri_random)
stri = vmi.ent.calc_anova(listMI_Anova, sufix)
# desk, speciesParams, roundVal=4, filename=None, stri = '', saveData=False):
vmi.ent.print_data_summary(desk, speciesParams, roundVal=roundVal, filename=filename, stri=stri)