def _get_ordered_files(self):
"""
get all permuted files and return it in a dict with as key phenotype and a list of filenames
"""
files = glob.glob(self.inout.out + "*.perm") #get perm *.perm
pattern = re.compile(self.inout.out + ".{1,8}\.P(?P<pheno>.*)\.perm")
files_clean = [x for x in files if pattern.search(x) != None]
if (len(files) < 1):
log.info("Could not find any permutations named as " + self.inout.out + "*.perm")
log.info(common.get_terminated_time())
sys.exit()
ordered_results = _get_ordered_files_from_list(files_clean)
return (ordered_results)
def mergeresults(self):
"""
function to merge multiple permutation files into one file
"""
#load sumlog files to create sumlog files
aa_result = self._load_sumlog_files()
ordered_results = self._get_ordered_results()
perm_files = self._get_ordered_files()
keys = ordered_results.keys()
keys.sort(key=common.alpha_sort)
pheno_nr = 0
#merge each phenotype
for key in keys:
pheno_nr=pheno_nr+1
log.info("\nMerging " + (str(len(perm_files[key]))) + " permutation files for phenotype " + key + "...")
perm_out_string = ordered_results[key].format_permout() # concatenated results
perm_filename = "merged.P" + key + ".perm"
perm_out_filename = self.inout.save_text_to_filename(perm_filename, perm_out_string)
log.info("Saved merged permutations as " + perm_out_filename)
#save empirical P file if sumlog files is found
if (aa_result.has_key(key)):
aa_object = Clusterresults()
aa_object.read_formated_results(common.getfile_handle(aa_result[key]))
empp_out_as_text = ordered_results[key].format_permutated_results(aa_object)
emp_filename = "merged.P" + key + ".empp"
empp_filename = self.inout.save_text_to_filename(emp_filename, empp_out_as_text)
log.info("Saved empirical pvalues as " + empp_filename)
#call R for distribution plot
self.files[key] = {"perm":perm_out_filename, "empp":empp_filename}
if self.inout.run_rproject:
import jag.plot_with_r as plot_with_r
plotter = plot_with_r.call_r(self.inout)
plotter.draw_dist_plot(self.files, key)
else:
log.info("\nWarning: Could not find sumlog file " + self.inout.out + ".P" + key + ".sumlog")
log.info(common.get_terminated_time())
sys.exit()
def format_permout(self):
"""format the permutations score in a table like manner with as column headers
groupname.
"""
groupnames = self.permutated_scores.keys()
groupnames.sort(key=common.alpha_sort)
length_permutations = [len(self.permutated_scores[g]) for g in groupnames]
diffrence_in_length = sum([abs(l - length_permutations[0])for l in length_permutations])
formated_results = '\t'.join(groupnames) + "\tseed\n"
if(diffrence_in_length == 0):
for i in range(length_permutations[0]):
formated_results += '\t'.join([str(self.permutated_scores[p][i])for p in groupnames])
formated_results += "\t" + str(self.seeds[i]) + "\n"
else:
log.info("\nWarning: gene-sets are not equal over all permuted files. Analysis will be terminated.")
log.info(common.get_terminated_time())
exit()
return(formated_results)