def pearson_correlation(datadir,
outdir,
usemask=True,
excludef='',
exclude_idx=-1):
slidir = datadir + os.path.sep + au.slices_str()
subjsfile = datadir + os.path.sep + au.subjects_str()
labelsfile = datadir + os.path.sep + au.labels_str()
lst = os.listdir(slidir)
n = au.count_match(lst, au.data_str() + '_' + au.slice_regex())
exclude_log = ''
if exclude_idx > -1:
exclude_log = ' excluding subject ' + str(exclude_idx)
au.log.info('Calculating correlation of ' + slidir + os.path.sep +
au.data_str() + '_' + au.slice_regex() + exclude_log)
for i in range(n):
slino = au.zeropad(i)
dataf = slidir + os.path.sep + au.data_str() + '_' + au.slice_str(
) + '_' + slino + au.ext_str()
maskf = slidir + os.path.sep + au.mask_str() + '_' + au.slice_str(
) + '_' + slino + au.ext_str()
outf = outdir + os.path.sep + au.pearson_str() + '_' + au.slice_str(
) + '_' + slino
if exclude_idx > -1:
outf += '_' + au.excluded_str() + str(exclude_idx) + au.ext_str()
else:
outf += au.ext_str()
if not os.path.isfile(dataf):
au.log.error('Could not find ' + dataf)
continue
if not usemask:
maskf = ''
try:
measure_pearson(dataf, labelsfile, outf, maskf, excludef,
exclude_idx)
except:
au.log.error(
'pearson_correlation: Error measuring correlation on ' + dataf)
au.log.error("Unexpected error: ", sys.exc_info()[0])
exit(1)
def pearson_correlation (datadir, outdir, usemask=True, excludef='', exclude_idx=-1):
slidir = datadir + os.path.sep + au.slices_str()
subjsfile = datadir + os.path.sep + au.subjects_str()
labelsfile = datadir + os.path.sep + au.labels_str()
lst = os.listdir(slidir)
n = au.count_match(lst, au.data_str() + '_' + au.slice_regex())
exclude_log = ''
if exclude_idx > -1:
exclude_log = ' excluding subject ' + str(exclude_idx)
au.log.info ('Calculating correlation of ' + slidir + os.path.sep + au.data_str() + '_' + au.slice_regex() + exclude_log)
for i in range(n):
slino = au.zeropad(i)
dataf = slidir + os.path.sep + au.data_str() + '_' + au.slice_str() + '_' + slino + au.ext_str()
maskf = slidir + os.path.sep + au.mask_str() + '_' + au.slice_str() + '_' + slino + au.ext_str()
outf = outdir + os.path.sep + au.pearson_str() + '_' + au.slice_str() + '_' + slino
if exclude_idx > -1:
outf += '_' + au.excluded_str() + str(exclude_idx) + au.ext_str()
else:
outf += au.ext_str()
if not os.path.isfile(dataf):
au.log.error('Could not find ' + dataf)
continue
if not usemask:
maskf = ''
try:
measure_pearson(dataf, labelsfile, outf, maskf, excludef, exclude_idx)
except:
au.log.error('pearson_correlation: Error measuring correlation on ' + dataf)
au.log.error("Unexpected error: ", sys.exc_info()[0] )
exit(1)
if (excluf):
outf_exclude = au.exclude_str()
if expname:
outf_exclude += '_' + expname
if foldnumber:
outf_exclude += '_' + foldnumber
np.savetxt(outdir + os.path.sep + outf_exclude, excluded, fmt='%i')
np.savetxt(mdir + os.path.sep + au.exclude_str(),
excluded,
fmt='%i')
excluf = mdir + os.path.sep + au.exclude_str()
step = au.maskmerging_str() + ' ' + measure_fname + ' ' + mdir
if usemask and not au.is_done(chkf, step):
maskregex = au.mask_str() + '_' + au.slice_str() + '*'
post.merge_slices(slidir, maskregex, au.mask_str(), mdir, False)
au.checklist_add(chkf, step)
#CORRELATION
#read the measure argument and start processing
if measure == 'pea':
#measure pearson correlation for each population slice
step = au.measureperslice_str() + step_params
if not au.is_done(chkf, step):
pear.pearson_correlation(outdir, mdir, usemask, excluf, leave)
au.checklist_add(chkf, step)
#merge all correlation slice measures
step = au.postmerging_str() + step_params
if not au.is_done(chkf, step):
#saving exclude files in mdir
outf_exclude = ''
if (excluf):
outf_exclude = au.exclude_str()
if expname:
outf_exclude += '_' + expname
if foldnumber:
outf_exclude += '_' + foldnumber
np.savetxt(outdir + os.path.sep + outf_exclude , excluded, fmt='%i')
np.savetxt(mdir + os.path.sep + au.exclude_str(), excluded, fmt='%i')
excluf = mdir + os.path.sep + au.exclude_str()
step = au.maskmerging_str() + ' ' + measure_fname + ' ' + mdir
if usemask and not au.is_done(chkf, step):
maskregex = au.mask_str() + '_' + au.slice_str() + '*'
post.merge_slices (slidir, maskregex, au.mask_str(), mdir, False)
au.checklist_add(chkf, step)
#CORRELATION
#read the measure argument and start processing
if measure == 'pea':
#measure pearson correlation for each population slice
step = au.measureperslice_str() + step_params
if not au.is_done(chkf, step):
pear.pearson_correlation (outdir, mdir, usemask, excluf, leave)
au.checklist_add(chkf, step)
#merge all correlation slice measures
step = au.postmerging_str() + step_params
if not au.is_done(chkf, step):
def slice_and_merge (subsfname, labelsfname, checkfname, outdir='', maskfname='', outbase=''):
check_data (subsfname, labelsfname, outdir, maskfname, outbase)
nsubs = au.file_len(subsfname)
nlabs = au.file_len(labelsfname)
if not outdir:
outdir = os.getcwd()
slicesdir = outdir + os.path.sep + au.slices_str()
tempdir = outdir + os.path.sep + au.temp_str()
tmpdirlst = os.listdir(tempdir)
fpath = ''
olddir = os.getcwd()
if not au.is_done (checkfname, au.preslicingdata_str()):
#slicing subjects
au.log.info ('Slicing all subjects: takes a while')
subsfile = open(subsfname, 'r')
for line in subsfile:
fpath = line.strip()
isfile = os.path.basename(fpath)
osfile = tempdir + os.path.sep + isfile
#test if file has been sliced
fdim3 = au.fslval(fpath, 'dim3')
regex = au.remove_ext(isfile) + '*'
nslices = au.count_match (tmpdirlst, regex)
if fdim3 != nslices:
#if not, then slice it in tempdir
au.log.debug('Slicing ' + isfile)
os.chdir (tempdir)
shutil.copy (fpath, tempdir)
au.fslslice (osfile)
os.remove(osfile)
os.chdir (olddir)
else:
au.log.debug(isfile + ' previously sliced')
subsfile.close()
au.checklist_add (checkfname, au.preslicingdata_str())
if not au.is_done (checkfname, au.premergingdata_str()):
#merging each slice
if not outbase:
outbase = au.data_str()
if not fpath:
subsfile = open(subsfname, 'r')
for line in subsfile:
fpath = line.strip()
break
au.log.info ('Merging all subject slices: takes a while')
nslices = int(au.fslval(fpath,'dim3'))
for slice in range(nslices):
slicezp = au.zeropad(slice)
mergeout = outbase + '_slice_' + slicezp
outdata = slicesdir + os.path.sep + mergeout
#check if data file exists
if not au.imtest(outdata):
#if not, then create it, merging all the corresponding slices
au.log.debug('Merging slice ' + slicezp)
imglob = ''
subsfile = open(subsfname, 'r')
for line in subsfile:
fpath = line.strip()
isfile = au.remove_ext(os.path.basename(fpath)).strip()
isfile = tempdir + os.path.sep + isfile + '_slice_' + slicezp
imglob += isfile + ' '
os.system('fslmerge -t ' + outdata + ' ' + imglob)
for f in imglob.split():
au.imrm(f)
else:
au.log.debug('Slice ' + slicezp + ' previously done')
au.checklist_add (checkfname, au.premergingdata_str())
#slicing mask
if maskfname:
if os.path.exists(maskfname) and not au.is_done (checkfname, au.preslicingmask_str()):
au.log.info('Slicing mask ' + maskfname)
au.imcp (maskfname, slicesdir + os.path.sep + au.mask_str())
os.chdir (slicesdir)
au.fslslice (au.mask_str())
au.imrm (au.mask_str())
au.checklist_add (checkfname, au.preslicingmask_str())
os.chdir (olddir)
au.log.debug('Done preprocessing')