def routine(gui, DATA_FILE, ANNOT_FILE):
# gui_thread = threading.Thread(target=gui.write_to_output, args=(gui, "Reading the file...\n",))
# gui_thread.start()
# gui_thread.join()
gui.write_to_output("Reading the file...\n")
grph = Graphics()
fromFile = readTheFile(DATA_FILE, gui)
alternatives = findAlternatives(fromFile)
gui.write_to_output("Total transcripts: " + str(len(fromFile)) +
" | APA: " + str(len(alternatives)) + "\n")
if len(alternatives) > 0:
gui.write_to_output("Found alternatives...\n")
else:
gui.write_to_output("No alternatives, check the arguments\n")
raise SystemExit
fracs = calculateFractions(alternatives)
# data = fracs[0].getSamples()
# symbols = [fracs[0].getName()]
# for frac in fracs[1:]:
# data = np.vstack((data, frac.getSamples()))
# symbols.append(frac.getName())
# pca = PCAVisual(data, symbols)
# pca.show(DATA_FILE)
gui.write_to_output("Read the annotations file...\n")
annotations = readAnnotation(ANNOT_FILE, gui)
gui.write_to_output("Checks the annotations...\n")
findAnnotatedShifts(fracs, annotations, gui)
shifts = findShifts(fracs)
shifts = correct_fdr(shifts)
temp = []
for gene in shifts:
if gene.getMaxShift() > 1.5:
temp.append(gene)
# grph.data_to_heat_map(temp, names, filename="str_above1d5.pdf")
topdf2 = grph.scatter_pval_to_fold(shifts, shift=1.5, gui=gui)
sorted(topdf2, key=lambda x: x.getName())
fm = SimpleMotifsFinder.Family()
sequences = open("utrs_all_alt.fa", 'w')
threads = []
alt_exon = check_cds(topdf2, annotations, gui)
long_up, long_down = grph.length_by_regulation(
[gene for gene in topdf2 if gene not in alt_exon], SAMPLES_PARTS)
# print([gene.getName() for gene in long_up])
# print([gene.getName() for gene in long_down])
"""
The next lines can be executed using multiprocessing or multithreading.
These are some times for each approach(run on set of 37 genes):
Threading: 50 seconds
Processing: 137.8 seconds
Neither: 57.1 seconds
"""
# for gene in fracs:
# seq = gene.getSequence()
# print(gene.getName())
# sequences.write(">" + gene.getName() + "\n")
# sequences.write(seq + "\n")
# thread = threading.Thread(target=fm.hash_sequence, args=(seq,))
# thread.start()
# threads.append(thread)
# # fm.hash_sequence(seq)
# # if gene.getName() == 'Camk2a':
# # grph.show_change_in_box(gene, SAMPLES_PARTS)
# for thread in threads:
# thread.join()
# sequences.close()
# fm.write_motifs()
# check_cds(topdf2, annotations)
grph.fold_change_and_pvalue(shifts)
gui.write_to_output("Writing the output...\n")
gui.write_to_output("Done, press 'Exit' to close the window.\n")
