def Run(self):
self.transit_message("Starting IGV Export")
start_time = time.time()
#Get orf data
self.transit_message("Getting Data")
(fulldata, position) = tnseq_tools.get_data(self.ctrldata)
(fulldata,
factors) = norm_tools.normalize_data(fulldata, self.normalization,
self.ctrldata,
self.annotation_path)
position = position.astype(int)
hash = transit_tools.get_pos_hash(self.annotation_path)
rv2info = transit_tools.get_gene_info(self.annotation_path)
self.transit_message("Normalizing")
self.output.write("#Converted to IGV with TRANSIT.\n")
if self.normalization != "nonorm":
self.output.write("#Reads normalized using '%s'\n" %
self.normalization)
if type(factors[0]) == type(0.0):
self.output.write(
"#Normalization Factors: %s\n" %
"\t".join(["%s" % f for f in factors.flatten()]))
else:
self.output.write("#Normalization Factors: %s\n" % " ".join(
[",".join(["%s" % bx for bx in b]) for b in factors]))
self.output.write("#Files:\n")
for f in self.ctrldata:
self.output.write("#%s\n" % f)
dataset_str = "\t".join(
[transit_tools.fetch_name(F) for F in self.ctrldata])
self.output.write("#Chromosome\tStart\tEnd\tFeature\t%s\tTAs\n" %
dataset_str)
chrom = transit_tools.fetch_name(self.annotation_path)
(K, N) = fulldata.shape
self.progress_range(N)
for i, pos in enumerate(position):
self.output.write(
"%s\t%s\t%s\tTA%s\t%s\t1\n" %
(chrom, position[i], position[i] + 1, position[i], "\t".join(
["%1.1f" % fulldata[j][i] for j in range(len(fulldata))])))
# Update progress
text = "Running Export Method... %5.1f%%" % (100.0 * i / N)
self.progress_update(text, i)
self.output.close()
self.transit_message("") # Printing empty line to flush stdout
self.finish()
self.transit_message("Finished Export")
def Run(self):
self.transit_message("Starting IGV Export")
start_time = time.time()
#Get orf data
self.transit_message("Getting Data")
(fulldata, position) = tnseq_tools.get_data(self.ctrldata)
(fulldata, factors) = norm_tools.normalize_data(fulldata, self.normalization,
self.ctrldata, self.annotation_path)
position = position.astype(int)
hash = transit_tools.get_pos_hash(self.annotation_path)
rv2info = transit_tools.get_gene_info(self.annotation_path)
self.transit_message("Normalizing")
self.output.write("#Converted to IGV with TRANSIT.\n")
if self.normalization != "nonorm":
self.output.write("#Reads normalized using '%s'\n" % self.normalization)
if type(factors[0]) == type(0.0):
self.output.write("#Normalization Factors: %s\n" % "\t".join(["%s" % f for f in factors.flatten()]))
else:
self.output.write("#Normalization Factors: %s\n" % " ".join([",".join(["%s" % bx for bx in b]) for b in factors]))
self.output.write("#Files:\n")
for f in self.ctrldata:
self.output.write("#%s\n" % f)
dataset_str = "\t".join([transit_tools.fetch_name(F) for F in self.ctrldata])
self.output.write("#Chromosome\tStart\tEnd\tFeature\t%s\tTAs\n" % dataset_str)
chrom = transit_tools.fetch_name(self.annotation_path)
(K,N) = fulldata.shape
self.progress_range(N)
for i,pos in enumerate(position):
self.output.write("%s\t%s\t%s\tTA%s\t%s\t1\n" % (chrom, position[i], position[i]+1, position[i], "\t".join(["%1.1f" % fulldata[j][i] for j in range(len(fulldata))])))
# Update progress
text = "Running Export Method... %5.1f%%" % (100.0*i/N)
self.progress_update(text, i)
self.output.close()
self.transit_message("") # Printing empty line to flush stdout
self.finish()
self.transit_message("Finished Export")
def displayHistogram(self, displayFrame, event):
gene = displayFrame.grid.GetCellValue(displayFrame.row, 0)
filepath = os.path.join(ntpath.dirname(displayFrame.path), transit_tools.fetch_name(displayFrame.path))
filename = os.path.join(filepath, gene+".png")
if os.path.exists(filename):
imgWindow = pytransit.fileDisplay.ImgFrame(None, filename)
imgWindow.Show()
else:
transit_tools.ShowError(MSG="Error Displaying File. Histogram image not found. Make sure results were obtained with the histogram option turned on.")
print("Error Displaying File. Histogram image does not exist.")
def displayHistogram(self, displayFrame, event):
gene = displayFrame.grid.GetCellValue(displayFrame.row, 0)
filepath = os.path.join(ntpath.dirname(displayFrame.path), transit_tools.fetch_name(displayFrame.path))
filename = os.path.join(filepath, gene+".png")
if os.path.exists(filename):
imgWindow = pytransit.fileDisplay.ImgFrame(None, filename)
imgWindow.Show()
else:
transit_tools.ShowError(MSG="Error Displaying File. Histogram image not found. Make sure results were obtained with the histogram option turned on.")
print "Error Displaying File. Histogram image does not exist."
def Run(self):
self.transit_message("Starting Gene Mean Counts Export")
start_time = time.time()
#Get orf data
self.transit_message("Getting Data")
(fulldata, position) = tnseq_tools.get_data(self.ctrldata)
(fulldata, factors) = norm_tools.normalize_data(fulldata, self.normalization,
self.ctrldata, self.annotation_path)
position = position.astype(int)
hash = transit_tools.get_pos_hash(self.annotation_path)
rv2info = transit_tools.get_gene_info(self.annotation_path)
self.transit_message("Normalizing")
self.output.write("#Summarized to Mean Gene Counts with TRANSIT.\n")
if self.normalization != "nonorm":
self.output.write("#Reads normalized using '%s'\n" % self.normalization)
if type(factors[0]) == type(0.0):
self.output.write("#Normalization Factors: %s\n" % "\t".join(["%s" % f for f in factors.flatten()]))
else:
self.output.write("#Normalization Factors: %s\n" % " ".join([",".join(["%s" % bx for bx in b]) for b in factors]))
self.output.write("#Files:\n")
for f in self.ctrldata:
self.output.write("#%s\n" % f)
K,Nsites = fulldata.shape
# Get Gene objects
G = tnseq_tools.Genes(self.ctrldata, self.annotation_path, norm=self.normalization)
N = len(G)
self.progress_range(N)
dataset_header = "\t".join([transit_tools.fetch_name(D) for D in self.ctrldata])
self.output.write("#Orf\tName\tNumber of TA sites\t%s\n" % dataset_header)
for i,gene in enumerate(G):
if gene.n > 0:
data_str = "\t".join(["%1.2f" % (M) for M in numpy.mean(gene.reads, 1)])
else:
data_str = "\t".join(["%1.2f" % (Z) for Z in numpy.zeros(K)])
self.output.write("%s\t%s\t%s\t%s\n" % (gene.orf, gene.name, gene.n, data_str))
# Update progress
text = "Running Export Method... %5.1f%%" % (100.0*i/N)
self.progress_update(text, i)
self.output.close()
self.transit_message("") # Printing empty line to flush stdout
self.finish()
self.transit_message("Finished Export")
def Run(self):
#if not self.wxobj:
# # Force matplotlib to use good backend for png.
# import matplotlib.pyplot as plt
#elif "matplotlib.pyplot" not in sys.modules:
try:
import matplotlib.pyplot as plt
except:
print "Error: cannot do histograms"
self.doHistogram = False
self.transit_message("Starting resampling Method")
start_time = time.time()
if self.doHistogram:
histPath = os.path.join(
os.path.dirname(self.output.name),
transit_tools.fetch_name(self.output.name) + "_histograms")
if not os.path.isdir(histPath):
os.makedirs(histPath)
else:
histPath = ""
Kctrl = len(self.ctrldata)
Kexp = len(self.expdata)
#Get orf data
self.transit_message("Getting Data")
(data, position) = transit_tools.get_validated_data(self.ctrldata +
self.expdata,
wxobj=self.wxobj)
(K, N) = data.shape
if self.normalization != "nonorm":
self.transit_message("Normalizing using: %s" % self.normalization)
(data,
factors) = norm_tools.normalize_data(data, self.normalization,
self.ctrldata + self.expdata,
self.annotation_path)
if self.LOESS:
self.transit_message("Performing LOESS Correction")
for j in range(K):
data[j] = stat_tools.loess_correction(position, data[j])
G = tnseq_tools.Genes(self.ctrldata + self.expdata,
self.annotation_path,
ignoreCodon=self.ignoreCodon,
nterm=self.NTerminus,
cterm=self.CTerminus,
data=data,
position=position)
#G = tnseq_tools.Genes(self.ctrldata+self.expdata, self.annotation_path, norm=self.normalization, ignoreCodon=self.ignoreCodon, nterm=self.NTerminus, cterm=self.CTerminus)
#Resampling
data = []
N = len(G)
count = 0
self.progress_range(N)
for gene in G:
count += 1
if gene.k == 0 or gene.n == 0:
(test_obs, mean1, mean2, log2FC, pval_ltail, pval_utail,
pval_2tail, testlist, data1, data2) = (0, 0, 0, 0, 1.00, 1.00,
1.00, [], [0], [0])
else:
if not self.includeZeros:
ii = numpy.sum(gene.reads, 0) > 0
else:
ii = numpy.ones(gene.n) == 1
data1 = gene.reads[:Kctrl, ii].flatten() + self.pseudocount
data2 = gene.reads[Kctrl:, ii].flatten() + self.pseudocount
(test_obs, mean1, mean2, log2FC, pval_ltail, pval_utail,
pval_2tail, testlist) = stat_tools.resampling(
data1,
data2,
S=self.samples,
testFunc=stat_tools.F_mean_diff_flat,
adaptive=self.adaptive)
if self.doHistogram:
import matplotlib.pyplot as plt
if testlist:
n, bins, patches = plt.hist(testlist,
density=1,
facecolor='c',
alpha=0.75,
bins=100)
else:
n, bins, patches = plt.hist([0, 0],
density=1,
facecolor='c',
alpha=0.75,
bins=100)
plt.xlabel('Delta Mean')
plt.ylabel('Probability')
plt.title('%s - Histogram of Delta Mean' % gene.orf)
plt.axvline(test_obs,
color='r',
linestyle='dashed',
linewidth=3)
plt.grid(True)
genePath = os.path.join(histPath, gene.orf + ".png")
if not os.path.exists(histPath):
os.makedirs(histPath)
plt.savefig(genePath)
plt.clf()
sum1 = numpy.sum(data1)
sum2 = numpy.sum(data2)
data.append([
gene.orf, gene.name, gene.desc, gene.n, mean1, mean2, sum1,
sum2, test_obs, log2FC, pval_2tail
])
# Update progress
text = "Running Resampling Method... %5.1f%%" % (100.0 * count / N)
self.progress_update(text, count)
#
self.transit_message("") # Printing empty line to flush stdout
self.transit_message("Performing Benjamini-Hochberg Correction")
data.sort()
qval = stat_tools.BH_fdr_correction([row[-1] for row in data])
self.output.write("#Resampling\n")
if self.wxobj:
members = sorted([
attr for attr in dir(self) if not callable(getattr(self, attr))
and not attr.startswith("__")
])
memberstr = ""
for m in members:
memberstr += "%s = %s, " % (m, getattr(self, m))
self.output.write(
"#GUI with: norm=%s, samples=%s, pseudocounts=%1.2f, adaptive=%s, histogram=%s, includeZeros=%s, output=%s\n"
% (self.normalization, self.samples, self.pseudocount,
self.adaptive, self.doHistogram, self.includeZeros,
self.output.name.encode('utf-8')))
else:
self.output.write("#Console: python %s\n" % " ".join(sys.argv))
self.output.write("#Control Data: %s\n" %
(",".join(self.ctrldata).encode('utf-8')))
self.output.write("#Experimental Data: %s\n" %
(",".join(self.expdata).encode('utf-8')))
self.output.write("#Annotation path: %s\n" %
(self.annotation_path.encode('utf-8')))
self.output.write("#Time: %s\n" % (time.time() - start_time))
self.output.write("#%s\n" % "\t".join(columns))
for i, row in enumerate(data):
(orf, name, desc, n, mean1, mean2, sum1, sum2, test_obs, log2FC,
pval_2tail) = row
self.output.write(
"%s\t%s\t%s\t%d\t%1.1f\t%1.1f\t%1.2f\t%1.1f\t%1.2f\t%1.1f\t%1.5f\t%1.5f\n"
% (orf, name, desc, n, mean1, mean2, log2FC, sum1, sum2,
test_obs, pval_2tail, qval[i]))
self.output.close()
self.transit_message("Adding File: %s" % (self.output.name))
self.add_file(filetype="Resampling")
self.finish()
self.transit_message("Finished resampling Method")
def Run(self):
#if not self.wxobj:
# # Force matplotlib to use good backend for png.
# import matplotlib.pyplot as plt
#elif "matplotlib.pyplot" not in sys.modules:
try:
import matplotlib.pyplot as plt
except:
print("Error: cannot do histograms")
self.doHistogram = False
self.transit_message("Starting resampling Method")
start_time = time.time()
histPath = ""
if self.doHistogram:
histPath = os.path.join(os.path.dirname(self.output.name), transit_tools.fetch_name(self.output.name)+"_histograms")
if not os.path.isdir(histPath):
os.makedirs(histPath)
#Get orf data
self.transit_message("Getting Data")
if self.diffStrains:
self.transit_message("Multiple annotation files found")
self.transit_message("Mapping ctrl data to {0}, exp data to {1}".format(self.annotation_path, self.annotation_path_exp))
if self.combinedWigParams:
(position, data, filenamesInCombWig) = tnseq_tools.read_combined_wig(self.combinedWigParams['combined_wig'])
conditionsByFile, _, _, _ = tnseq_tools.read_samples_metadata(self.combinedWigParams['samples_metadata'])
conditions = self.wigs_to_conditions(conditionsByFile, filenamesInCombWig)
data, conditions = self.filter_wigs_by_conditions(data, conditions, self.combinedWigParams['conditions'])
data_ctrl = numpy.array([d for i, d in enumerate(data) if conditions[i].lower() == self.combinedWigParams['conditions'][0]])
data_exp = numpy.array([d for i, d in enumerate(data) if conditions[i].lower() == self.combinedWigParams['conditions'][1]])
position_ctrl, position_exp = position, position
else:
(data_ctrl, position_ctrl) = transit_tools.get_validated_data(self.ctrldata, wxobj=self.wxobj)
(data_exp, position_exp) = transit_tools.get_validated_data(self.expdata, wxobj=self.wxobj)
(K_ctrl, N_ctrl) = data_ctrl.shape
(K_exp, N_exp) = data_exp.shape
if not self.diffStrains and (N_ctrl != N_exp):
self.transit_error("Error: Ctrl and Exp wig files don't have the same number of sites.")
self.transit_error("Make sure all .wig files come from the same strain.")
return
# (data, position) = transit_tools.get_validated_data(self.ctrldata+self.expdata, wxobj=self.wxobj)
self.transit_message("Preprocessing Ctrl data...")
data_ctrl = self.preprocess_data(position_ctrl, data_ctrl)
self.transit_message("Preprocessing Exp data...")
data_exp = self.preprocess_data(position_exp, data_exp)
G_ctrl = tnseq_tools.Genes(self.ctrldata, self.annotation_path, ignoreCodon=self.ignoreCodon, nterm=self.NTerminus, cterm=self.CTerminus, data=data_ctrl, position=position_ctrl)
G_exp = tnseq_tools.Genes(self.expdata, self.annotation_path_exp, ignoreCodon=self.ignoreCodon, nterm=self.NTerminus, cterm=self.CTerminus, data=data_exp, position=position_exp)
doLibraryResampling = False
# If library string not empty
if self.ctrl_lib_str or self.exp_lib_str:
letters_ctrl = set(self.ctrl_lib_str)
letters_exp = set(self.exp_lib_str)
# Check if using exactly 1 letters; i.e. no different libraries
if len(letters_ctrl) == 1 and letters_exp==1:
pass
# If using more than one letter, then check no differences in set
else:
lib_diff = letters_ctrl ^ letters_exp
# Check that their differences
if not lib_diff:
doLibraryResampling = True
else:
transit_tools.transit_error("Error: Library Strings (Ctrl = %s, Exp = %s) do not use the same letters. Make sure every letter / library is represented in both Control and Experimental Conditions. Proceeding with resampling assuming all datasets belong to the same library." % (self.ctrl_lib_str, self.exp_lib_str))
self.ctrl_lib_str = ""
self.exp_lib_str = ""
(data, qval) = self.run_resampling(G_ctrl, G_exp, doLibraryResampling, histPath)
self.write_output(data, qval, start_time)
self.finish()
self.transit_message("Finished resampling Method")
def Run(self):
#if not self.wxobj:
# # Force matplotlib to use good backend for png.
# import matplotlib.pyplot as plt
#elif "matplotlib.pyplot" not in sys.modules:
try:
import matplotlib.pyplot as plt
except:
print "Error: cannot do histograms"
self.doHistogram = False
self.transit_message("Starting resampling Method")
start_time = time.time()
histPath = ""
if self.doHistogram:
histPath = os.path.join(os.path.dirname(self.output.name), transit_tools.fetch_name(self.output.name)+"_histograms")
if not os.path.isdir(histPath):
os.makedirs(histPath)
#Get orf data
self.transit_message("Getting Data")
if self.diffStrains:
self.transit_message("Multiple annotation files found")
self.transit_message("Mapping ctrl data to {0}, exp data to {1}".format(self.annotation_path, self.annotation_path_exp))
if self.combinedWigParams:
(position, data, filenamesInCombWig) = tnseq_tools.read_combined_wig(self.combinedWigParams['combined_wig'])
conditionsByFile, _, _, _ = tnseq_tools.read_samples_metadata(self.combinedWigParams['samples_metadata'])
conditions = self.wigs_to_conditions(conditionsByFile, filenamesInCombWig)
data, conditions = self.filter_wigs_by_conditions(data, conditions, self.combinedWigParams['conditions'])
data_ctrl = numpy.array([d for i, d in enumerate(data) if conditions[i].lower() == self.combinedWigParams['conditions'][0]])
data_exp = numpy.array([d for i, d in enumerate(data) if conditions[i].lower() == self.combinedWigParams['conditions'][1]])
position_ctrl, position_exp = position, position
else:
(data_ctrl, position_ctrl) = transit_tools.get_validated_data(self.ctrldata, wxobj=self.wxobj)
(data_exp, position_exp) = transit_tools.get_validated_data(self.expdata, wxobj=self.wxobj)
(K_ctrl, N_ctrl) = data_ctrl.shape
(K_exp, N_exp) = data_exp.shape
if not self.diffStrains and (N_ctrl != N_exp):
self.transit_error("Error: Ctrl and Exp wig files don't have the same number of sites.")
self.transit_error("Make sure all .wig files come from the same strain.")
return
# (data, position) = transit_tools.get_validated_data(self.ctrldata+self.expdata, wxobj=self.wxobj)
self.transit_message("Preprocessing Ctrl data...")
data_ctrl = self.preprocess_data(position_ctrl, data_ctrl)
self.transit_message("Preprocessing Exp data...")
data_exp = self.preprocess_data(position_exp, data_exp)
G_ctrl = tnseq_tools.Genes(self.ctrldata, self.annotation_path, ignoreCodon=self.ignoreCodon, nterm=self.NTerminus, cterm=self.CTerminus, data=data_ctrl, position=position_ctrl)
G_exp = tnseq_tools.Genes(self.expdata, self.annotation_path_exp, ignoreCodon=self.ignoreCodon, nterm=self.NTerminus, cterm=self.CTerminus, data=data_exp, position=position_exp)
doLibraryResampling = False
# If library string not empty
if self.ctrl_lib_str or self.exp_lib_str:
letters_ctrl = set(self.ctrl_lib_str)
letters_exp = set(self.exp_lib_str)
# Check if using exactly 1 letters; i.e. no different libraries
if len(letters_ctrl) == 1 and letters_exp==1:
pass
# If using more than one letter, then check no differences in set
else:
lib_diff = letters_ctrl ^ letters_exp
# Check that their differences
if not lib_diff:
doLibraryResampling = True
else:
transit_tools.transit_error("Error: Library Strings (Ctrl = %s, Exp = %s) do not use the same letters. Make sure every letter / library is represented in both Control and Experimental Conditions. Proceeding with resampling assuming all datasets belong to the same library." % (self.ctrl_lib_str, self.exp_lib_str))
self.ctrl_lib_str = ""
self.exp_lib_str = ""
(data, qval) = self.run_resampling(G_ctrl, G_exp, doLibraryResampling, histPath)
self.write_output(data, qval, start_time)
self.finish()
self.transit_message("Finished resampling Method")
