def save_results(self, output_dir='.', prefix='', prefix_sep='_',
image_list=None):
""" Write out any images generated by the meta-analysis.
Args:
output_dir (str): folder to write images to
prefix (str): all image files will be prepended with this string
prefix_sep (str): glue between the prefix and rest of filename
image_list (list): optional list of images to save--e.g.,
['pFgA_z', 'pAgF']. If image_list is None (default), will save
all images.
"""
if prefix == '':
prefix_sep = ''
if not exists(output_dir):
makedirs(output_dir)
logger.debug("Saving results...")
if image_list is None:
image_list = self.images.keys()
for suffix, img in self.images.items():
if suffix in image_list:
filename = prefix + prefix_sep + suffix + '.nii.gz'
outpath = join(output_dir, filename)
imageutils.save_img(img, outpath, self.dataset.masker)
def make_mask_map(data, infile, outfile, index=None):
from neurosynth.base.mask import Masker
from neurosynth.base import imageutils
# Load image with masker
masker = Masker(infile)
img = imageutils.load_imgs(infile, masker)
data = list(data)
if index is None:
index = np.arange(0, len(data))
rev_index = None
else:
all_reg = np.arange(0, img.max())
rev_index = all_reg[np.invert(np.in1d(all_reg, index))]
min_val = img.min()
for num, value in enumerate(data):
n = index[num]
np.place(img, img == n + min_val, [value])
if rev_index is not None:
for value in rev_index:
np.place(img, img == value + min_val, 0)
img = img.astype('float32')
imageutils.save_img(img, outfile, masker)
def make_mask_map(data, infile, outfile, index=None):
from neurosynth.base.mask import Masker
from neurosynth.base import imageutils
# Load image with masker
masker = Masker(infile)
img = imageutils.load_imgs(infile, masker)
img = np.round(img)
data = list(data)
if index is None:
index = np.unique(img)
rev_index = None
else:
all_reg = np.arange(0, img.max())
rev_index = all_reg[np.invert(np.in1d(all_reg, index))]
min_val = img.min()
for num, value in enumerate(data):
ix = index[num]
np.place(img, img == ix, [value])
if rev_index is not None:
for value in rev_index:
np.place(img, img == value + min_val, 0)
img = img.astype('float32')
imageutils.save_img(img, outfile, masker)
def _create_cluster_images(self, labels, output_dir=None):
''' Creates a Nifti image of reconstructed cluster labels.
Args:
labels: A vector of cluster labels
output_dir: A string indicating folder to output images to. If None,
creates a "ClusterImages" directory below the Clusterer instance's
output directory.
Outputs:
Cluster_k.nii.gz: Will output a nifti image with cluster labels
'''
labels += 1
# Reconstruct grid into original space
# TODO: replace with masker.unmask()
if hasattr(self, 'grid'):
regions = self.masker.mask(self.grid)
unique_regions = np.unique(regions)
n_regions = unique_regions.size
m = np.zeros(regions.size)
for i in range(n_regions):
m[regions == unique_regions[i]] = labels[i] + 1
labels = m
if output_dir is None:
output_dir = os.path.join(self.output_dir, 'ClusterImages')
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
outfile = os.path.join(output_dir,'Cluster_k%d.nii.gz' % (len(np.unique(labels))))
imageutils.save_img(labels, outfile, self.masker)
def save_results(self,
output_dir='.',
prefix='',
prefix_sep='_',
image_list=None):
""" Write out any images generated by the meta-analysis.
Args:
output_dir (str): folder to write images to
prefix (str): all image files will be prepended with this string
prefix_sep (str): glue between the prefix and rest of filename
image_list (list): optional list of images to save--e.g.,
['pFgA_z', 'pAgF']. If image_list is None (default), will save
all images.
"""
if prefix == '':
prefix_sep = ''
if not exists(output_dir):
makedirs(output_dir)
logger.debug("Saving results...")
if image_list is None:
image_list = self.images.keys()
for suffix, img in self.images.items():
if suffix in image_list:
filename = prefix + prefix_sep + suffix + '.nii.gz'
outpath = join(output_dir, filename)
imageutils.save_img(img, outpath, self.dataset.masker)
def dimension_reduction(self, reducer, n_components=100, save=False):
""" Reduces the dimensionalty of the currently loaded reference data.
Args:
reducer (str or sklearn object): The reduction method to use. Can be
either a string (one of 'pca', 'ica', or 'ward') or an object that
follows sklearn's TransformerMixin pattern.
n_components: Number of components to extract, if applicable. When
method is an object, this argument is ignored.
save: if True, writes out images of the components.
"""
if isinstance(reducer, basestring):
_valid = {
'pca': RandomizedPCA,
'ica': FastICA,
}
if reducer not in _valid:
raise ValueError("Dimensionality reduction method must be one "
"of %s; '%s' is not a valid value." %
(str(_valid.keys), reducer))
reducer = _valid[reducer](n_components=n_components)
self.reference_data = reducer.fit_transform(self.reference_data.T).T
if True:
print "Saving components..."
for i in range(n_components):
comp = reducer.components_[i, :]
outf = join(self.output_dir, 'ICA_component_%d.nii.gz' % i)
print comp.min(), comp.max(), comp.dtype, type(comp), comp.shape
imageutils.save_img((comp-comp.mean())/comp.std(), outf, self.masker)
def dimension_reduction(self, reducer, n_components=100, save=False):
""" Reduces the dimensionalty of the currently loaded reference data.
Args:
reducer (str or sklearn object): The reduction method to use. Can be
either a string (one of 'pca', 'ica', or 'ward') or an object that
follows sklearn's TransformerMixin pattern.
n_components: Number of components to extract, if applicable. When
method is an object, this argument is ignored.
save: if True, writes out images of the components.
"""
if isinstance(reducer, basestring):
_valid = {
'pca': RandomizedPCA,
'ica': FastICA,
}
if reducer not in _valid:
raise ValueError("Dimensionality reduction method must be one "
"of %s; '%s' is not a valid value." %
(str(_valid.keys), reducer))
reducer = _valid[reducer](n_components=n_components)
self.reference_data = reducer.fit_transform(self.reference_data.T).T
if True:
print "Saving components..."
for i in range(n_components):
comp = reducer.components_[i, :]
outf = join(self.output_dir, 'ICA_component_%d.nii.gz' % i)
print comp.min(), comp.max(), comp.dtype, type(
comp), comp.shape
imageutils.save_img((comp - comp.mean()) / comp.std(), outf,
self.masker)
def _create_cluster_images(self, labels, output_dir=None):
''' Creates a Nifti image of reconstructed cluster labels.
Args:
labels: A vector of cluster labels
output_dir: A string indicating folder to output images to. If None,
creates a "ClusterImages" directory below the Clusterer instance's
output directory.
Outputs:
Cluster_k.nii.gz: Will output a nifti image with cluster labels
'''
labels += 1
# Reconstruct grid into original space
# TODO: replace with masker.unmask()
if hasattr(self, 'grid'):
regions = self.masker.mask(self.grid)
unique_regions = np.unique(regions)
n_regions = unique_regions.size
m = np.zeros(regions.size)
for i in range(n_regions):
m[regions == unique_regions[i]] = labels[i] + 1
labels = m
if output_dir is None:
output_dir = os.path.join(self.output_dir, 'ClusterImages')
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
outfile = os.path.join(output_dir,
'Cluster_k%d.nii.gz' % (len(np.unique(labels))))
imageutils.save_img(labels, outfile, self.masker)
def save_results(self, outroot, image_list=None):
""" Write out any images generated by the meta-analysis. The outroot argument is prepended
to all file names. Optionally, a restricted list of images to save can be passed; otherwise,
all images currently stored in self.images will be saved. """
print "Saving results..."
if image_list == None: image_list = self.images.keys()
for suffix, img in self.images.items():
if suffix in image_list:
imageutils.save_img(img, '%s_%s.nii.gz' % (outroot, suffix), self.dataset.volume)
def neurosynthMatch(db,papers,author,outdir=None,outprefix=None):
"""Match neurosynth id with papers id"""
# Get all IDs in neuroSynth
neurosynth_ids = getIDs(db)
# LIST OF IDS ---------------------------------------------------------
# Input is DOI with list of papers
if bool(re.search("[/]",papers[0])):
# NeuroSynth is also DOI
if bool(re.search("[/]",neurosynth_ids[0])):
print "Search for " + str(len(papers)) + " ids in NeuroSynth database..."
# Find intersection
valid_ids = [x for x in papers if x in neurosynth_ids]
# Neurosynth is PMID
else:
print "ERROR: Please provide doi to use the 525 database!"
sys.exit()
# Input is pmid with list of papers
else:
# NeuroSynth is also pmid
if not bool(re.search("[/]",neurosynth_ids[0])):
print "Search for " + str(len(papers)) + " ids in NeuroSynth database..."
# Find intersection
valid_ids = [x for x in papers if x in neurosynth_ids]
# Neurosynth is doi
else:
print "ERROR: Please provide pmid to use the 3000 database!"
sys.exit()
if (len(valid_ids) > 0):
# Do meta analysis
ma = meta.MetaAnalysis(db,valid_ids)
# 1) the z score map corresponding to the probability that a study in the database is tagged with a particular feature given that activation is present at a particular voxel, FDR corrected .05
dataFDR = ma.images[ma.images.keys()[1]]
# 2) the probability of feature given activation with uniform prior imposed
dataPRIOR = ma.images[ma.images.keys()[6]]
# 3) the probability of feature given activation
data = ma.images[ma.images.keys()[7]]
# 4) the probability of feature given activation, Z score
dataZ = ma.images[ma.images.keys()[8]]
# 5) z score map corresponding to the probability of activation given that a study is tagged with the feature (author)
datapAgF = ma.images[ma.images.keys()[4]]
# If user specifies an output directory
if outdir:
print "Saving results to output directory " + outdir + "..."
if not outprefix:
outprefix = author.replace(" ","")
imageutils.save_img(datapAgF, '%s/%s_pAgF_z_FDR_0.05.nii.gz' % (outdir, outprefix), db.volume)
imageutils.save_img(dataFDR, '%s/%s_pFgA_z_FDR_0.05.nii.gz' % (outdir, outprefix), db.volume)
imageutils.save_img(dataPRIOR, '%s/%s_pFgA_given_pF=0.50.nii.gz' % (outdir, outprefix), db.volume)
imageutils.save_img(data, '%s/%s_pFgA.nii.gz' % (outdir, outprefix), db.volume)
imageutils.save_img(dataZ, '%s/%s_pFgA_z.nii.gz' % (outdir, outprefix), db.volume)
return ma.images
else:
print "No overlapping studies found in database for author " + author + "."
def remove_value(infile, vals_rm, outfile):
masker = mask.Masker(infile)
img = imageutils.load_imgs(infile, masker)
# Remove value
for val in vals_rm:
np.place(img, img == val, [0])
# Save
imageutils.save_img(img, outfile, masker)
def save_results(self, outroot, image_list=None):
""" Write out any images generated by the meta-analysis. The outroot argument is prepended
to all file names. Optionally, a restricted list of images to save can be passed; otherwise,
all images currently stored in self.images will be saved. """
print "Saving results..."
if image_list == None: image_list = self.images.keys()
for suffix, img in self.images.items():
if suffix in image_list:
imageutils.save_img(img, '%s_%s.nii.gz' % (outroot, suffix),
self.dataset.volume)
def remove_value(infile, vals_rm, outfile):
masker = mask.Masker(infile)
img = imageutils.load_imgs(infile, masker)
img = np.round(img)
# Remove value
for val in vals_rm:
np.place(img, img == val, [0])
# Save
imageutils.save_img(img, outfile, masker)
def _create_cluster_images(self, labels, coactivation_maps):
''' Creates a Nifti image of reconstructed cluster labels.
Args:
labels: A vector of cluster labels
Outputs:
Cluster_k.nii.gz: Will output a nifti image with cluster labels
'''
# Reconstruct grid into original space
# TODO: replace with masker.unmask()
if hasattr(self, 'grid'):
regions = self.masker.mask(self.grid)
unique_regions = np.unique(regions)
n_regions = unique_regions.size
m = np.zeros(regions.size)
for i in range(n_regions):
m[regions == unique_regions[i]] = labels[i] + 1
labels = m
clusters = np.unique(labels)
n_clusters = len(clusters)
prefix = '' if self.prefix is None else self.prefix + '_'
output_dir = join(self.output_dir,
prefix + self.algorithm + '_k' + str(n_clusters))
if not isdir(output_dir):
os.makedirs(output_dir)
outfile = join(output_dir, 'cluster_labels.nii.gz')
imageutils.save_img(labels, outfile, self.masker)
# Generate a coactivation map for each cluster
if coactivation_maps:
coact_dir = join(output_dir, 'coactivation')
if not isdir(coact_dir):
os.makedirs(coact_dir)
for c in clusters:
img = np.zeros_like(labels)
img[labels == c] = 1
img = self.masker.unmask(img)
ids = self.dataset.get_ids_by_mask(img, 0.25)
ma = meta.MetaAnalysis(self.dataset, ids)
ma.save_results(coact_dir, 'cluster_%d' % c)
def _create_cluster_images(self, labels, coactivation_maps):
''' Creates a Nifti image of reconstructed cluster labels.
Args:
labels: A vector of cluster labels
Outputs:
Cluster_k.nii.gz: Will output a nifti image with cluster labels
'''
# Reconstruct grid into original space
# TODO: replace with masker.unmask()
if hasattr(self, 'grid'):
regions = self.masker.mask(self.grid)
unique_regions = np.unique(regions)
n_regions = unique_regions.size
m = np.zeros(regions.size)
for i in range(n_regions):
m[regions == unique_regions[i]] = labels[i] + 1
labels = m
clusters = np.unique(labels)
n_clusters = len(clusters)
prefix = '' if self.prefix is None else self.prefix + '_'
output_dir = join(self.output_dir, prefix + self.algorithm + '_k' + str(n_clusters))
if not isdir(output_dir):
os.makedirs(output_dir)
outfile = join(output_dir,prefix + self.algorithm + '_k' + str(n_clusters) + 'cluster_labels.nii.gz')
imageutils.save_img(labels, outfile, self.masker)
# Generate a coactivation map for each cluster
if coactivation_maps:
coact_dir = join(output_dir, 'coactivation')
if not isdir(coact_dir):
os.makedirs(coact_dir)
for c in clusters:
img = np.zeros_like(labels)
img[labels==c] = 1
img = self.masker.unmask(img)
ids = self.dataset.get_ids_by_mask(img, 0.25)
ma = meta.MetaAnalysis(self.dataset, ids)
ma.save_results(coact_dir, 'cluster_%d' % c)
def neurosynthQuery(searchTerm,thresh,dataset,outdir=None):
thresh = float(thresh)
query = dataset.get_ids_by_features('*' + searchTerm + '*',threshold=thresh)
ma = meta.MetaAnalysis(dataset,query)
# This gets the absolute value FDR corrected at threshold
# 'pAgF_z_FDR_0.05'
# the z score map corresponding to the map of the probability of activation given that a study is tagged with the feature, FDR corrected .05
data = ma.images[ma.images.keys()[4]]
# Print this image to file, to look at later
if outdir:
imageutils.save_img(data, '%s/%s.nii.gz' % (outdir, searchTerm), dataset.volume)
img = dataset.volume.unmask(data)
# These are x,y,z coordinates of nonnzero voxels
idx = np.nonzero(img)
affine = dataset.volume.volume.get_affine()
coords = np.dot(affine,[ idx[0],idx[1],idx[2] ,np.ones(len(idx[0]))])
coords = np.transpose(coords)
# Get rid of 4th column
coords = np.delete(coords,-1,1)
# These are MNI coordinates for non-zero voxels
return coords
def neurosynthMatch(db, papers, author, outdir=None, outprefix=None):
"""Match neurosynth id with papers id"""
# Get all IDs in neuroSynth
neurosynth_ids = getIDs(db)
# LIST OF IDS ---------------------------------------------------------
# Input is DOI with list of papers
if bool(re.search("[/]", papers[0])):
# NeuroSynth is also DOI
if bool(re.search("[/]", neurosynth_ids[0])):
print "Search for " + str(
len(papers)) + " ids in NeuroSynth database..."
# Find intersection
valid_ids = [x for x in papers if x in neurosynth_ids]
# Neurosynth is PMID
else:
print "ERROR: Please provide doi to use the 525 database!"
sys.exit()
# Input is pmid with list of papers
else:
# NeuroSynth is also pmid
if not bool(re.search("[/]", neurosynth_ids[0])):
print "Search for " + str(
len(papers)) + " ids in NeuroSynth database..."
# Find intersection
valid_ids = [x for x in papers if x in neurosynth_ids]
# Neurosynth is doi
else:
print "ERROR: Please provide pmid to use the 3000 database!"
sys.exit()
if (len(valid_ids) > 0):
# Do meta analysis
ma = meta.MetaAnalysis(db, valid_ids)
# 1) the z score map corresponding to the probability that a study in the database is tagged with a particular feature given that activation is present at a particular voxel, FDR corrected .05
dataFDR = ma.images[ma.images.keys()[1]]
# 2) the probability of feature given activation with uniform prior imposed
dataPRIOR = ma.images[ma.images.keys()[6]]
# 3) the probability of feature given activation
data = ma.images[ma.images.keys()[7]]
# 4) the probability of feature given activation, Z score
dataZ = ma.images[ma.images.keys()[8]]
# 5) z score map corresponding to the probability of activation given that a study is tagged with the feature (author)
datapAgF = ma.images[ma.images.keys()[4]]
# If user specifies an output directory
if outdir:
print "Saving results to output directory " + outdir + "..."
if not outprefix:
outprefix = author.replace(" ", "")
imageutils.save_img(
datapAgF, '%s/%s_pAgF_z_FDR_0.05.nii.gz' % (outdir, outprefix),
db.volume)
imageutils.save_img(
dataFDR, '%s/%s_pFgA_z_FDR_0.05.nii.gz' % (outdir, outprefix),
db.volume)
imageutils.save_img(
dataPRIOR,
'%s/%s_pFgA_given_pF=0.50.nii.gz' % (outdir, outprefix),
db.volume)
imageutils.save_img(data,
'%s/%s_pFgA.nii.gz' % (outdir, outprefix),
db.volume)
imageutils.save_img(dataZ,
'%s/%s_pFgA_z.nii.gz' % (outdir, outprefix),
db.volume)
return ma.images
else:
print "No overlapping studies found in database for author " + author + "."
from neurosynth.base.dataset import Dataset
import neurosynth.base.imageutils as it
dataset = Dataset.load("../data/datasets/abs_topics_filt.pkl")
print "Filtering voxels..."
data = dataset.image_table.data.toarray()
voxel_mask = data.mean(axis=1) > 0.005
img = it.load_imgs('../masks/ward/30.nii.gz', dataset.masker)
good_voxels = img[voxel_mask]
it.save_img(good_voxels, "../masks/ward/30_masked.nii.gz", dataset.masker)
dataset = Dataset.load("../data/datasets/abs_topics_filt.pkl")
print "Filtering voxels..."
data = dataset.image_table.data.toarray()
voxel_mask = data.mean(axis=1) > 0.0135
good_voxels = data[voxel_mask]
good_voxels = sparse.csr_matrix(good_voxels)
for i in [20, 30, 40, 50]:
print "Clustering..."
print i
k_means = KMeans(init='k-means++', n_clusters=i, n_jobs=16)
k_means.fit(good_voxels)
# ward = Ward(n_clusters=30)
# ward.fit(good_voxels)
print "Stretching clustering results..."
cluster_voxels = np.zeros(voxel_mask.shape)
cluster_voxels[voxel_mask] = k_means.labels_ + 1
print "Saving image..."
save_img(cluster_voxels, "k_means_" + str(i) + "_0135.nii.gz", dataset.masker)
# Remove those not in a good community
bad = list(set(zero_ix))
# Remove
for value in bad:
np.place(img, img == value, [0])
non_0_ix.sort()
# Translate numbers to continous
translated = zip(range(1, non_0_ix.shape[0]+1), list(non_0_ix))
for pair in translated:
np.place(img, img == pair[1], pair[0])
print "New shape:"
print np.bincount([int(vox) for vox in img]).shape
imageutils.save_img(img, outfile, masker)
# Write key
with open('../masks/Andy/parcels_' + str(min_vox) + '_key.csv', 'w') as file:
writer = csv.writer(file)
writer.writerow(['new', 'original'])
for pair in translated:
writer.writerow(list(pair))
dataset = Dataset.load("../data/datasets/abs_topics_filt.pkl")
print "Filtering voxels..."
data = dataset.image_table.data.toarray()
voxel_mask = data.mean(axis=1) > 0.0135
good_voxels = data[voxel_mask]
good_voxels = sparse.csr_matrix(good_voxels)
for i in [20, 30, 40, 50]:
print "Clustering..."
print i
k_means = KMeans(init='k-means++', n_clusters=i, n_jobs=16)
k_means.fit(good_voxels)
# ward = Ward(n_clusters=30)
# ward.fit(good_voxels)
print "Stretching clustering results..."
cluster_voxels = np.zeros(voxel_mask.shape)
cluster_voxels[voxel_mask] = k_means.labels_ + 1
print "Saving image..."
save_img(cluster_voxels, "k_means_" + str(i) + "_0135.nii.gz",
dataset.masker)