def test_twosample(self):
data, vardata, XYZ = make_data(n=20)
data1,vardata1,data2,vardata2 = data[:10],vardata[:10],data[10:],vardata[10:]
# rfx calibration
P = PT.permutation_test_twosample(data1, data2, XYZ, ndraws=ndraws)
c = [(P.random_Tvalues[P.ndraws*(0.95)],None),(P.random_Tvalues[P.ndraws*(0.5)],10)]
r = [np.zeros(data.shape[1])]
r[data.shape[1]/2:] *= 10
#p_values, cluster_results, region_results=P.calibrate(nperms=100, clusters=c, regions=r)
# mfx calibration
P = PT.permutation_test_twosample(data1, data2, XYZ, vardata1=vardata1, vardata2=vardata2, stat_id="student_mfx", ndraws=ndraws)
p_values, cluster_results, region_results = P.calibrate(nperms=nperms, clusters=c, regions=r)
def twosample_test(data_images, vardata_images, mask_images, labels, stat_id, permutations=0, cluster_forming_th=0.01):
"""
Helper function for permutation-based mass univariate twosample group
analysis. Labels is a binary vector (1-2). Regions more active for group
1 than group 2 are inferred.
"""
# Prepare arrays
data, vardata, xyz, mask = prepare_arrays(data_images, vardata_images, mask_images)
# Create two-sample permutation test instance
if vardata_images == None:
ptest = permutation_test_twosample(data[labels == 1], data[labels == 2], xyz, stat_id=stat_id)
else:
ptest = permutation_test_twosample(
data[labels == 1],
data[labels == 2],
xyz,
vardata1=vardata[labels == 1],
vardata2=vardata[labels == 2],
stat_id=stat_id,
)
# Compute z-map image
zmap = np.zeros(data_images[0].get_shape()).squeeze()
zmap[list(xyz)] = ptest.zscore()
zimg = Image(zmap, data_images[0].get_affine())
# Compute mask image
maskimg = Image(mask, data_images[0].get_affine())
# Multiple comparisons
if permutations <= 0:
return zimg, maskimg
else:
# Cluster definition: (threshold, diameter)
cluster_def = (ptest.height_threshold(cluster_forming_th), None)
# Calibration
voxel_res, cluster_res, region_res = ptest.calibrate(nperms=permutations, clusters=[cluster_def])
nulls = {}
nulls["zmax"] = ptest.zscore(voxel_res["perm_maxT_values"])
nulls["s"] = cluster_res[0]["perm_size_values"]
nulls["smax"] = cluster_res[0]["perm_maxsize_values"]
# Return z-map image, mask image and dictionary of null
# distribution for cluster sizes (s), max cluster size (smax)
# and max z-score (zmax)
return zimg, maskimg, nulls
r = np.ones(data.shape[1], int)
r[data.shape[1]/2.:] *= 10
voxel_results, cluster_results, region_results = \
P.calibrate(nperms=100, clusters=c, regions=[r])
# mfx calibration
P = PT.permutation_test_onesample(data, XYZ, vardata=vardata,
stat_id="student_mfx")
voxel_results, cluster_results, region_results = \
P.calibrate(nperms=100, clusters=c, regions=[r])
################################################################################
# Example for using permutation_test_twosample class
data, vardata, XYZ = make_data(n=20)
data1, vardata1, data2, vardata2 = data[:10], vardata[:10], data[10:], vardata[10:]
# rfx calibration
P = PT.permutation_test_twosample(data1,data2,XYZ)
# clusters definition (height threshold / max diameter)
c = [(P.random_Tvalues[P.ndraws * (0.95)], None),
(P.random_Tvalues[P.ndraws * (0.5)], 10)]
# regions definition (label vector)
r = [np.zeros(data.shape[1])]
r[data.shape[1]/2:] *= 10
voxel_results, cluster_results, region_results = \
P.calibrate(nperms=100, clusters=c, regions = r)
# mfx calibration
P = PT.permutation_test_twosample(data1, data2, XYZ, vardata1=vardata1,
vardata2=vardata2, stat_id="student_mfx")
voxel_results, cluster_results, region_results = \
P.calibrate(nperms=100, clusters=c, regions=r)
################################################################################
