def Run(self):
self.transit_message("Starting Griffin Method")
start_time = time.time()
#Get orf data
self.transit_message("Getting Data")
(data, position) = transit_tools.get_validated_data(self.ctrldata,
wxobj=self.wxobj)
(K, N) = data.shape
if self.normalization and self.normalization != "nonorm":
self.transit_message("Normalizing using: %s" % self.normalization)
(data,
factors) = norm_tools.normalize_data(data, self.normalization,
self.ctrldata,
self.annotation_path)
G = tnseq_tools.Genes(self.ctrldata,
self.annotation_path,
minread=1,
reps=self.replicates,
ignoreCodon=self.ignoreCodon,
nterm=self.NTerminus,
cterm=self.CTerminus,
data=data,
position=position)
N = len(G)
self.progress_range(N)
count = 0
pins = G.global_theta()
pnon = 1.0 - pins
results = []
for gene in G:
if gene.n == 0:
results.append([gene, 0.0, 1.000])
else:
B = 1.0 / math.log(1.0 / pnon)
u = math.log(gene.n * pins, 1.0 / pnon)
exprun = tnseq_tools.ExpectedRuns(gene.n, pnon)
pval = 1.0 - tnseq_tools.GumbelCDF(gene.r, u, B)
results.append([gene, exprun, pval])
text = "Running Griffin Method... %5.1f%%" % (100.0 * (count + 1) /
(N))
self.progress_update(text, count)
count += 1
pval = [row[-1] for row in results]
padj = stat_tools.BH_fdr_correction(pval)
for i in range(len(results)):
results[i].append(padj[i])
results.sort()
self.output.write("#Griffin\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: ctrldata=%s, annotation=%s, output=%s\n" %
(",".join(self.ctrldata).encode('utf-8'),
self.annotation_path.encode('utf-8'),
self.output.name.encode('utf-8')))
else:
self.output.write("#Console: python3 %s\n" % " ".join(sys.argv))
self.output.write("#Data: %s\n" %
(",".join(self.ctrldata).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 (gene, exprun, pval, padj) in results:
self.output.write(
"%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\t%1.1f\t%1.5f\t%1.5f\n" %
(gene.orf, gene.name, gene.desc, gene.k, gene.n, gene.r,
gene.s, gene.t, exprun, pval, padj))
self.output.close()
self.transit_message("") # Printing empty line to flush stdout
self.transit_message("Adding File: %s" % (self.output.name))
self.add_file(filetype="Griffin")
self.finish()
self.transit_message("Finished Griffin Method")
def Run(self):
self.transit_message("Starting Tn5 gaps method")
start_time = time.time()
self.transit_message("Getting data (May take a while)")
# Combine all wigs
(data, position) = transit_tools.get_validated_data(self.ctrldata,
wxobj=self.wxobj)
combined = tnseq_tools.combine_replicates(data, method=self.replicates)
combined[combined < self.minread] = 0
counts = combined
counts[counts > 0] = 1
num_sites = counts.size
genes_obj = tnseq_tools.Genes(self.ctrldata,
self.annotation_path,
ignoreCodon=self.ignoreCodon,
nterm=self.NTerminus,
cterm=self.CTerminus,
data=data,
position=position)
pins = numpy.mean(counts)
pnon = 1.0 - pins
# Calculate stats of runs
exprunmax = tnseq_tools.ExpectedRuns(num_sites, pnon)
varrun = tnseq_tools.VarR(num_sites, pnon)
stddevrun = math.sqrt(varrun)
exp_cutoff = exprunmax + 2 * stddevrun
# Get the runs
self.transit_message("Getting non-insertion runs in genome")
run_arr = tnseq_tools.runs_w_info(counts)
pos_hash = transit_tools.get_pos_hash(self.annotation_path)
# Finally, calculate the results
self.transit_message("Running Tn5 gaps method")
results_per_gene = {}
for gene in genes_obj.genes:
results_per_gene[gene.orf] = [
gene.orf, gene.name, gene.desc, gene.k, gene.n, gene.r, 0, 0, 1
]
N = len(run_arr)
count = 0
accum = 0
self.progress_range(N)
for run in run_arr:
accum += run['length']
count += 1
genes = tnseq_tools.get_genes_in_range(pos_hash, run['start'],
run['end'])
for gene_orf in genes:
gene = genes_obj[gene_orf]
inter_sz = self.intersect_size([run['start'], run['end']],
[gene.start, gene.end]) + 1
percent_overlap = self.calc_overlap([run['start'], run['end']],
[gene.start, gene.end])
run_len = run['length']
B = 1.0 / math.log(1.0 / pnon)
u = math.log(num_sites * pins, 1.0 / pnon)
pval = 1.0 - tnseq_tools.GumbelCDF(run['length'], u, B)
curr_val = results_per_gene[gene.orf]
curr_inter_sz = curr_val[6]
curr_len = curr_val[7]
if inter_sz > curr_inter_sz:
results_per_gene[gene.orf] = [
gene.orf, gene.name, gene.desc, gene.k, gene.n, gene.r,
inter_sz, run_len, pval
]
# Update Progress
text = "Running Tn5Gaps method... %1.1f%%" % (100.0 * count / N)
self.progress_update(text, count)
data = list(results_per_gene.values())
exp_run_len = float(accum) / N
min_sig_len = float('inf')
sig_genes_count = 0
pval = [row[-1] for row in data]
padj = stat_tools.BH_fdr_correction(pval)
for i in range(len(data)):
if padj[i] < 0.05:
sig_genes_count += 1
min_sig_len = min(min_sig_len, data[i][-2])
data[i].append(padj[i])
data[i].append('Essential' if padj[i] < 0.05 else 'Non-essential')
#(data[i][0], data[i][1], data[i][2], data[i][3], data[i][4], data[i][5], data[i][6], data[i][7], data[i][8], padj[i], 'Essential' if padj[i] < 0.05 else 'Non-essential')
data.sort(key=lambda l: l[0])
# Output results
self.output.write("#Tn5 Gaps\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: ctrldata=%s, annotation=%s, output=%s\n" %
(",".join(self.ctrldata).encode('utf-8'),
self.annotation_path.encode('utf-8'),
self.output.name.encode('utf-8')))
else:
self.output.write("#Console: python %s\n" % " ".join(sys.argv))
self.output.write("#Data: %s\n" %
(",".join(self.ctrldata).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("#Essential gene count: %d\n" % (sig_genes_count))
self.output.write("#Minimum reads: %d\n" % (self.minread))
self.output.write("#Replicate combination method: %s\n" %
(self.replicates))
self.output.write("#Minimum significant run length: %d\n" %
(min_sig_len))
self.output.write("#Expected run length: %1.5f\n" % (exp_run_len))
self.output.write("#Expected max run length: %s\n" % (exprunmax))
self.output.write("#%s\n" % "\t".join(columns))
#self.output.write("#Orf\tName\tDesc\tk\tn\tr\tovr\tlenovr\tpval\tpadj\tcall\n")
for res in data:
self.output.write(
"%s\t%s\t%s\t%s\t%s\t%s\t%d\t%d\t%1.5f\t%1.5f\t%s\n" %
(res[0], res[1], res[2], res[3], res[4], res[5], res[6],
res[7], res[8], res[9], res[10]))
self.output.close()
self.transit_message("") # Printing empty line to flush stdout
self.transit_message("Adding File: %s" % (self.output.name))
self.add_file(filetype="Tn5 Gaps")
self.finish()
self.transit_message("Finished Tn5Gaps Method")