def vtx_corr_col(data_file1, atlas_name, roi_name, data_file2, column, idx_col,
out_file):
"""
计算data_file1中指定atlas ROI内的所有顶点序列和data_file2中指定column序列的相关
Args:
data_file1 (str): end with .dscalar.nii
shape=(N, LR_count_32k)
atlas_name (str):
roi_name (str):
data_file2 (str): end with .csv
shape=(N, n_col)
column (str):
idx_col (int): specify index column of csv file
If None, means no index column.
"""
# prepare
reader = CiftiReader(data_file1)
maps = reader.get_data()
atlas = Atlas(atlas_name)
assert atlas.maps.shape == (1, LR_count_32k)
roi_idx_map = atlas.maps[0] == atlas.roi2label[roi_name]
maps = maps[:, roi_idx_map]
df = pd.read_csv(data_file2, index_col=idx_col)
col_vec = np.array(df[column])
col_vec = np.expand_dims(col_vec, 0)
# calculate
data = np.ones((1, LR_count_32k)) * np.nan
data[0, roi_idx_map] = 1 - cdist(col_vec, maps.T, 'correlation')[0]
# save
save2cifti(out_file, data, reader.brain_models())
def extract_net_parcel_info():
"""
把每个网络的编号，名字以及所包含的MMP parcel的编号和名字列出来
"""
from magicbox.io.io import CiftiReader
# inputs
net_file = pjoin(work_dir, 'networks.dlabel.nii')
# outputs
out_file = pjoin(work_dir, 'net_parcel_info.txt')
# prepare
reader = CiftiReader(net_file)
net_names = reader.map_names()
lbl_tables = reader.label_tables()
# calculate & save
wf = open(out_file, 'w')
for net_idx, net_name in enumerate(net_names):
wf.write(f'>>>{net_idx+1}-{net_name}\n')
lbl_tab = lbl_tables[net_idx]
for lbl_k, lbl_v in lbl_tab.items():
if lbl_k == 0:
continue
wf.write(f'{lbl_k}-{lbl_v.label[:-4]}\n')
wf.write('<<<\n')
wf.close()
def calc_meas_individual(hemi='lh'):
import nibabel as nib
import numpy as np
import pickle as pkl
from magicbox.io.io import CiftiReader
from cxy_hcp_ffa.lib.predefine import hemi2stru, roi2label
# inputs
work_dir = pjoin(proj_dir,
'analysis/s2/1080_fROI/refined_with_Kevin/tfMRI')
rois_file = pjoin(proj_dir, 'analysis/s2/1080_fROI/'
f'refined_with_Kevin/rois_v3_{hemi}.nii.gz')
meas_file = pjoin(proj_dir, 'analysis/s2/activation.dscalar.nii')
# outputs
out_file = pjoin(work_dir, f'individual_activ_{hemi}.pkl')
rois = nib.load(rois_file).get_data().squeeze().T
n_roi = len(roi2label)
meas_reader = CiftiReader(meas_file)
meas = meas_reader.get_data(hemi2stru[hemi], True)
n_subj = meas.shape[0]
roi_meas = {'shape': 'n_roi x n_subj', 'roi': list(roi2label.keys()),
'meas': np.ones((n_roi, n_subj)) * np.nan}
for roi_idx, roi in enumerate(roi_meas['roi']):
lbl_idx_arr = rois == roi2label[roi]
for subj_idx in range(n_subj):
lbl_idx_vec = lbl_idx_arr[subj_idx]
if np.any(lbl_idx_vec):
roi_meas['meas'][roi_idx, subj_idx] = np.mean(
meas[subj_idx][lbl_idx_vec])
pkl.dump(roi_meas, open(out_file, 'wb'))
def make_age_maps(data_file, info_file, out_name):
"""
对每个顶点，计算跨同一年龄被试的平均和sem，分别保存在
out_name-mean.dscalar.nii, out_name-sem.dscalar.nii中
Args:
data_file (str): end with .dscalar.nii
shape=(n_subj, LR_count_32k)
info_file (str): subject info file
out_name (str): filename to save
"""
# prepare
data_maps = nib.load(data_file).get_fdata()
info_df = pd.read_csv(info_file)
ages = np.array(info_df['age in years'])
ages_uniq = np.unique(ages)
n_age = len(ages_uniq)
# calculate
mean_maps = np.ones((n_age, LR_count_32k)) * np.nan
sem_maps = np.ones((n_age, LR_count_32k)) * np.nan
for age_idx, age in enumerate(ages_uniq):
data = data_maps[ages == age]
mean_maps[age_idx] = np.mean(data, 0)
sem_maps[age_idx] = sem(data, 0)
# save
map_names = [str(i) for i in ages_uniq]
reader = CiftiReader(mmp_map_file)
save2cifti(f'{out_name}-mean.dscalar.nii', mean_maps,
reader.brain_models(), map_names)
save2cifti(f'{out_name}-sem.dscalar.nii', sem_maps, reader.brain_models(),
map_names)
def merge_smoothed_data(dataset_name, meas_name, sigma):
"""
合并我平滑过后的cifti文件
Args:
dataset_name (str): HCPD | HCPA
meas_name (str): thickness | myelin
sigma (float): the size of the gaussian surface smoothing kernel in mm
"""
# outputs
out_file = pjoin(work_dir,
f'{dataset_name}_{meas_name}_{sigma}mm.dscalar.nii')
# prepare
src_dir = pjoin(work_dir, f'{dataset_name}_{meas_name}_{sigma}mm')
src_file = pjoin(src_dir, '{sid}.dscalar.nii')
df = pd.read_csv(dataset_name2info[dataset_name])
n_subj = df.shape[0]
data = np.zeros((n_subj, LR_count_32k), np.float64)
# calculate
for subj_idx, subj_id in enumerate(df['subID']):
time1 = time.time()
meas_file = src_file.format(sid=subj_id)
data[subj_idx] = nib.load(meas_file).get_fdata()[0]
print(f'Finished: {subj_idx+1}/{n_subj},'
f'cost: {time.time() - time1} seconds.')
# save
mmp_reader = CiftiReader(mmp_map_file)
save2cifti(out_file, data, mmp_reader.brain_models(), df['subID'])
def calc_meas_test(roi_type):
"""
用45个被试在test或retest中定出的个体FFA去提取各自在test的face selectivity。
Args:
roi_type (str): ROI-test or ROI-retest
"""
import numpy as np
import pandas as pd
import nibabel as nib
from magicbox.io.io import CiftiReader
from cxy_hcp_ffa.lib.predefine import hemi2stru, roi2label
# inputs
hemis = ('lh', 'rh')
rois = ('pFus-face', 'mFus-face')
subj_file_45 = pjoin(proj_dir, 'data/HCP/wm/analysis_s2/'
'retest/subject_id')
subj_file_1080 = pjoin(proj_dir, 'analysis/s2/subject_id')
meas_file = pjoin(proj_dir, 'analysis/s2/activation.dscalar.nii')
roi_type2file = {
'ROI-test':
pjoin(
proj_dir, 'analysis/s2/1080_fROI/refined_with_Kevin/'
'rois_v3_{hemi}.nii.gz'),
'ROI-retest':
pjoin(
proj_dir, 'analysis/s2/1080_fROI/refined_with_Kevin/'
'retest/rois_{hemi}_v2.nii.gz')
}
roi_file = roi_type2file[roi_type]
# outputs
out_file = pjoin(work_dir, f'activ-test_{roi_type}.csv')
# prepare
subj_ids_45 = open(subj_file_45).read().splitlines()
subj_ids_1080 = open(subj_file_1080).read().splitlines()
retest_idx_in_1080 = [subj_ids_1080.index(i) for i in subj_ids_45]
n_subj = len(subj_ids_45)
out_df = pd.DataFrame(index=range(n_subj))
# calculate
meas_reader = CiftiReader(meas_file)
for hemi in hemis:
meas_maps = meas_reader.get_data(hemi2stru[hemi],
True)[retest_idx_in_1080]
roi_maps = nib.load(roi_file.format(hemi=hemi)).get_fdata().squeeze().T
if roi_type == 'ROI-test':
roi_maps = roi_maps[retest_idx_in_1080]
for roi in rois:
col = f"{hemi}_{roi.split('-')[0]}"
for idx in range(n_subj):
roi_idx_map = roi_maps[idx] == roi2label[roi]
if np.any(roi_idx_map):
out_df.loc[idx, col] = np.mean(meas_maps[idx][roi_idx_map])
# save
out_df.to_csv(out_file, index=False)
def make_non_outlier_map(fpath,
thr,
roi_name,
out_file_mask=None,
out_file_prob=None):
"""
将同时在thr%以上的被试中被认定为outlier的顶点判定为跨被试的outlier
If out_file_mask is not None, save mask map. (.dlabel.nii or .npy)
If out_file_prob is not None, save prob map. (.dscalar.nii)
"""
# prepare
data = np.load(fpath)
n_subj, n_vtx = data.shape
atlas1 = Atlas('Cole_visual_LR')
atlas2 = Atlas('Cole_visual_ROI')
assert atlas1.maps.shape == (1, LR_count_32k)
assert atlas2.maps.shape == (1, LR_count_32k)
roi_idx_map = atlas1.maps[0] == atlas1.roi2label[roi_name]
if roi_name == 'R_cole_visual':
prefix = 'R_'
elif roi_name == 'L_cole_visual':
prefix = 'L_'
else:
raise ValueError("error roi_name:", roi_name)
mmp_reader = CiftiReader(mmp_map_file)
mmp_lbl_tab = mmp_reader.label_tables()[0]
# calculate
if out_file_mask is not None:
data_tmp = np.sum(data, axis=0)
outlier_vec = data_tmp > thr / 100 * n_subj
print(f'#outliers/total: {np.sum(outlier_vec)}/{n_vtx}')
mask_npy = np.zeros(LR_count_32k, bool)
mask_npy[roi_idx_map] = ~outlier_vec
if out_file_mask.endswith('.npy'):
np.save(out_file_mask, mask_npy)
elif out_file_mask.endswith('.dlabel.nii'):
mask_cii = atlas2.maps.copy()
mask_cii[0, ~mask_npy] = np.nan
lbl_tab = nib.cifti2.cifti2.Cifti2LabelTable()
for roi, lbl in atlas2.roi2label.items():
if roi.startswith(prefix):
lbl_tab[lbl] = mmp_lbl_tab[lbl]
save2cifti(out_file_mask,
mask_cii,
mmp_reader.brain_models(),
label_tables=[lbl_tab])
else:
raise ValueError("Not supported file name:", out_file_mask)
if out_file_prob is not None:
data_tmp = np.mean(data, axis=0)
prob_map = np.ones((1, LR_count_32k), dtype=np.float64) * np.nan
prob_map[0, roi_idx_map] = data_tmp
assert out_file_prob.endswith('.dscalar.nii')
save2cifti(out_file_prob, prob_map, mmp_reader.brain_models())
def separate_networks():
"""
把ColeNetwork的12个网络分到单独的map里。
每个map的MMP parcel的label保留原来的样子。
需要注意的是multimodal_glasser的MMP_mpmLR32k.dlabel.nii中，
ROI label的编号1~180是右脑，181~360是左脑。0对应的是???
而cortex_parcel_network_assignments.mat中0~359的index是左脑在先
"""
import numpy as np
import nibabel as nib
from scipy.io import loadmat
from cxy_visual_dev.lib.ColeNet import get_name_label_of_ColeNetwork
from magicbox.io.io import CiftiReader, save2cifti
# inputs
mmp_file = '/nfs/p1/atlases/multimodal_glasser/surface/'\
'MMP_mpmLR32k.dlabel.nii'
roi2net_file = pjoin(cole_dir, 'cortex_parcel_network_assignments.mat')
# outputs
out_file = pjoin(work_dir, 'networks.dlabel.nii')
# load
mmp_reader = CiftiReader(mmp_file)
mmp_map = mmp_reader.get_data()[0]
lbl_tab_raw = mmp_reader.label_tables()[0]
roi2net = loadmat(roi2net_file)['netassignments'][:, 0]
roi2net = np.r_[roi2net[180:], roi2net[:180]]
net_labels = np.unique(roi2net)
# prepare
data = np.zeros((len(net_labels), len(mmp_map)), dtype=np.uint16)
map_names = []
label_tables = []
net_lbl2name = {}
for name, lbl in zip(*get_name_label_of_ColeNetwork()):
net_lbl2name[lbl] = name
# calculate
for net_idx, net_lbl in enumerate(net_labels):
roi_labels = np.where(roi2net == net_lbl)[0] + 1
lbl_tab = nib.cifti2.cifti2.Cifti2LabelTable()
lbl_tab[0] = lbl_tab_raw[0]
for roi_lbl in roi_labels:
data[net_idx, mmp_map == roi_lbl] = roi_lbl
lbl_tab[roi_lbl] = lbl_tab_raw[roi_lbl]
map_names.append(net_lbl2name[net_lbl])
label_tables.append(lbl_tab)
# save
save2cifti(out_file,
data,
mmp_reader.brain_models(),
map_names,
label_tables=label_tables)
def calc_meas_individual(hemi='lh', meas='thickness'):
import nibabel as nib
import numpy as np
import pickle as pkl
from magicbox.io.io import CiftiReader
from cxy_hcp_ffa.lib.predefine import hemi2stru, roi2label
rois = ('IOG-face', 'pFus-face', 'mFus-face')
rois_file = pjoin(
proj_dir, 'analysis/s2/1080_fROI/'
f'refined_with_Kevin/rois_v3_{hemi}.nii.gz')
subj_file = pjoin(proj_dir, 'analysis/s2/subject_id')
meas_id_file = pjoin(proj_dir, 'data/HCP/subject_id_1096')
meas2file = {
'thickness':
'/nfs/p1/public_dataset/datasets/hcp/DATA/'
'HCP_S1200_GroupAvg_v1/HCP_S1200_GroupAvg_v1/'
'S1200.All.thickness_MSMAll.32k_fs_LR.dscalar.nii',
'myelin':
'/nfs/p1/public_dataset/datasets/hcp/DATA/'
'HCP_S1200_GroupAvg_v1/HCP_S1200_GroupAvg_v1/'
'S1200.All.MyelinMap_BC_MSMAll.32k_fs_LR.dscalar.nii',
'va':
'/nfs/p1/public_dataset/datasets/hcp/DATA/'
'HCP_S1200_GroupAvg_v1/HCP_S1200_GroupAvg_v1/'
'S1200.All.midthickness_MSMAll_va.32k_fs_LR.dscalar.nii'
}
meas_file = meas2file[meas]
trg_file = pjoin(work_dir, f'rois_v3_{hemi}_{meas}.pkl')
roi_maps = nib.load(rois_file).get_fdata().squeeze().T
n_roi = len(rois)
subj_ids = open(subj_file).read().splitlines()
n_subj = len(subj_ids)
meas_reader = CiftiReader(meas_file)
meas_ids = open(meas_id_file).read().splitlines()
meas_indices = [meas_ids.index(i) for i in subj_ids]
meas_maps = meas_reader.get_data(hemi2stru[hemi], True)[meas_indices]
roi_meas = {
'shape': 'n_roi x n_subj',
'roi': rois,
'meas': np.ones((n_roi, n_subj)) * np.nan
}
for roi_idx, roi in enumerate(roi_meas['roi']):
lbl_idx_arr = roi_maps == roi2label[roi]
for subj_idx in range(n_subj):
lbl_idx_vec = lbl_idx_arr[subj_idx]
meas_map = meas_maps[subj_idx]
if np.any(lbl_idx_vec):
if meas == 'va':
tmp = np.sum(meas_map[lbl_idx_vec])
else:
tmp = np.mean(meas_map[lbl_idx_vec])
roi_meas['meas'][roi_idx, subj_idx] = tmp
pkl.dump(roi_meas, open(trg_file, 'wb'))
def calc_pattern_corr_between_twins(meas_name='thickness'):
"""
Calculate spatial pattern correlation between each twin pair.
"""
import pandas as pd
import nibabel as nib
from magicbox.io.io import CiftiReader
from scipy.stats import pearsonr
twins_id_file = pjoin(work_dir, 'twins_id_1080.csv')
meas2file = {
'thickness':
'/nfs/p1/public_dataset/datasets/hcp/DATA/'
'HCP_S1200_GroupAvg_v1/HCP_S1200_GroupAvg_v1/'
'S1200.All.thickness_MSMAll.32k_fs_LR.dscalar.nii',
'myelin':
'/nfs/p1/public_dataset/datasets/hcp/DATA/'
'HCP_S1200_GroupAvg_v1/HCP_S1200_GroupAvg_v1/'
'S1200.All.MyelinMap_BC_MSMAll.32k_fs_LR.dscalar.nii',
'activ':
pjoin(proj_dir, 'analysis/s2/activation.dscalar.nii')
}
hemis = ('lh', 'rh')
hemi2stru = {
'lh': 'CIFTI_STRUCTURE_CORTEX_LEFT',
'rh': 'CIFTI_STRUCTURE_CORTEX_RIGHT'
}
mpm_file = pjoin(
proj_dir, 'analysis/s2/1080_fROI/refined_with_Kevin/'
'MPM_v3_{hemi}_0.25.nii.gz')
roi2label = {'IOG-face': 1, 'pFus-face': 2, 'mFus-face': 3}
zyg2label = {'MZ': 1, 'DZ': 2}
out_file = pjoin(work_dir, f'twins_pattern-corr_{meas_name}.csv')
df = pd.read_csv(twins_id_file)
meas_reader = CiftiReader(meas2file[meas_name])
meas_ids = [int(name.split('_')[0]) for name in meas_reader.map_names()]
twin1_indices = [meas_ids.index(i) for i in df['twin1']]
twin2_indices = [meas_ids.index(i) for i in df['twin2']]
out_df = pd.DataFrame()
out_df['zygosity'] = df['zygosity']
out_df['zyg'] = [zyg2label[zyg] for zyg in df['zygosity']]
for hemi in hemis:
meas1 = meas_reader.get_data(hemi2stru[hemi], True)[twin1_indices]
meas2 = meas_reader.get_data(hemi2stru[hemi], True)[twin2_indices]
mpm = nib.load(mpm_file.format(hemi=hemi)).get_data().squeeze()
for roi, lbl in roi2label.items():
idx_vec = mpm == lbl
out_df[f"{hemi}_{roi.split('-')[0]}"] = [
pearsonr(i[idx_vec], j[idx_vec])[0]
for i, j in zip(meas1, meas2)
]
out_df.to_csv(out_file, index=False)
def zscore_data(data_file, out_file):
"""
对每个被试做全脑zscore
Args:
data_file (str): .dscalar.nii
out_file (str): .dscalar.nii
"""
reader = CiftiReader(data_file)
data = reader.get_data()
data = zscore(data, 1)
save2cifti(out_file, data, reader.brain_models(), reader.map_names())
def check_1096_sid():
"""
检查HCPY_SubjInfo.csv文件中的1096个被试ID及其顺序和
S1200 1096 myelin file, S1200 1096 thickness file
是对的上的
"""
df = pd.read_csv(dataset_name2info['HCPY'])
subj_ids = df['subID'].to_list()
fpaths = (s1200_1096_myelin, s1200_1096_thickness)
for fpath in fpaths:
reader = CiftiReader(fpath)
tmp = [int(i.split('_')[0]) for i in reader.map_names()]
assert tmp == subj_ids
def calc_map_corr(data_file1,
data_file2,
atlas_name,
roi_name,
out_file,
map_names1=None,
map_names2=None,
index=False):
"""
计算指定atlas的ROI内的data1 maps和data2 maps的相关
存出的CSV文件的index是map_names1, column是map_names2
Args:
data_file1 (str): end with .dscalar.nii
shape=(n_map1, LR_count_32k)
data_file2 (str): end with .dscalar.nii
shape=(n_map2, LR_count_32k)
atlas_name (str):
roi_name (str):
out_file (str):
map_names1 (list, optional): Defaults to None.
If is None, use map names in data1_file.
map_names2 (list, optional): Defaults to None.
If is None, use map names in data2_file.
index (bool, optional): Defaults to False.
Save index of DataFrame or not
"""
# prepare
reader1 = CiftiReader(data_file1)
reader2 = CiftiReader(data_file2)
data_maps1 = reader1.get_data()
data_maps2 = reader2.get_data()
atlas = Atlas(atlas_name)
assert atlas.maps.shape == (1, LR_count_32k)
roi_idx_map = atlas.maps[0] == atlas.roi2label[roi_name]
data_maps1 = data_maps1[:, roi_idx_map]
data_maps2 = data_maps2[:, roi_idx_map]
if map_names1 is None:
map_names1 = reader1.map_names()
else:
assert len(map_names1) == data_maps1.shape[0]
if map_names2 is None:
map_names2 = reader2.map_names()
else:
assert len(map_names2) == data_maps2.shape[0]
# calculate
corrs = 1 - cdist(data_maps1, data_maps2, 'correlation')
# save
df = pd.DataFrame(corrs, map_names1, map_names2)
df.to_csv(out_file, index=index)
def pca(data_file, atlas_name, roi_name, n_component, axis, out_name):
"""
对n_subj x n_vtx形状的矩阵进行PCA降维
Args:
data_file (str): end with .dscalar.nii
shape=(n_subj, LR_count_32k)
atlas_name (str): include ROIs' labels and mask map
roi_name (str): 决定选用哪个区域内的顶点来参与PCA
n_component (int): the number of components
axis (str): vertex | subject
vertex: 对顶点数量进行降维，得到几个主成分时间序列，
观察某个主成分在各顶点上的权重，刻画其空间分布。
subject: 对被试数量进行降维，得到几个主成分map，
观察某个主成分在各被试上的权重，按年龄排序即可得到时间序列。
out_name (str): output name
If axis=vertex, output
1. n_subj x n_component out_name.csv
2. n_component x LR_count_32k out_name.dscalar.nii
3. out_name.pkl with fitted PCA model
"""
# prepare
component_names = [f'C{i}' for i in range(1, n_component + 1)]
meas_maps = nib.load(data_file).get_fdata()
atlas = Atlas(atlas_name)
assert atlas.maps.shape == (1, LR_count_32k)
roi_idx_map = atlas.maps[0] == atlas.roi2label[roi_name]
meas_maps = meas_maps[:, roi_idx_map]
bms = CiftiReader(mmp_map_file).brain_models()
# calculate
pca = PCA(n_components=n_component)
data = np.ones((n_component, LR_count_32k), np.float64) * np.nan
if axis == 'vertex':
X = meas_maps
pca.fit(X)
Y = pca.transform(X)
df = pd.DataFrame(data=Y, columns=component_names)
data[:, roi_idx_map] = pca.components_
elif axis == 'subject':
X = meas_maps.T
pca.fit(X)
Y = pca.transform(X)
df = pd.DataFrame(data=pca.components_.T, columns=component_names)
data[:, roi_idx_map] = Y.T
else:
raise ValueError('Invalid axis:', axis)
# save
df.to_csv(f'{out_name}.csv', index=False)
save2cifti(f'{out_name}.dscalar.nii', data, bms, component_names)
pkl.dump(pca, open(f'{out_name}.pkl', 'wb'))
def calc_meas_MPM(hemi='lh'):
"""
用一半被试的MPM去提取另一半被试的激活值
如果某个被试没有某个ROI，就不提取该被试该ROI的信号
Args:
hemi (str, optional): hemisphere. Defaults to 'lh'.
"""
import numpy as np
import pickle as pkl
import nibabel as nib
from cxy_hcp_ffa.lib.predefine import hemi2stru, roi2label
from magicbox.io.io import CiftiReader
hids = (1, 2)
hid_file = pjoin(split_dir, 'half_id.npy')
mpm_file = pjoin(split_dir, 'MPM_half{hid}_{hemi}.nii.gz')
roi_file = pjoin(proj_dir, 'analysis/s2/1080_fROI/refined_with_Kevin/'
f'rois_v3_{hemi}.nii.gz')
src_file = pjoin(proj_dir, 'analysis/s2/activation.dscalar.nii')
trg_file = pjoin(work_dir, f'activ_{hemi}.pkl')
hid_vec = np.load(hid_file)
n_subj = len(hid_vec)
roi_maps = nib.load(roi_file).get_data().squeeze().T
n_roi = len(roi2label)
meas_reader = CiftiReader(src_file)
meas = meas_reader.get_data(hemi2stru[hemi], True)
out_dict = {'shape': 'n_roi x n_subj',
'roi': list(roi2label.keys()),
'meas': np.ones((n_roi, n_subj)) * np.nan}
for hid in hids:
hid_idx_vec = hid_vec == hid
mpm = nib.load(mpm_file.format(hid=hid, hemi=hemi)
).get_data().squeeze()
for roi_idx, roi in enumerate(out_dict['roi']):
roi_idx_vec = np.any(roi_maps == roi2label[roi], 1)
valid_idx_vec = np.logical_and(~hid_idx_vec, roi_idx_vec)
mpm_idx_vec = mpm == roi2label[roi]
meas_masked = meas[valid_idx_vec][:, mpm_idx_vec]
out_dict['meas'][roi_idx][valid_idx_vec] = np.mean(meas_masked, 1)
pkl.dump(out_dict, open(trg_file, 'wb'))
def polyfit(data_file, info_file, deg, out_file):
"""
对时间序列进行多项式拟合
Args:
data_file (str): .csv | .dscalar.nii
If is .csv, fit each column with ages.
If is .dscalar.nii, fit each vertex with ages.
info_file (str): .csv
deg (int): degree of polynomial
out_file (str): file to output
same postfix with data_file
"""
# prepare
if data_file.endswith('.csv'):
assert out_file.endswith('.csv')
reader = pd.read_csv(data_file)
data = np.array(reader)
elif data_file.endswith('.dscalar.nii'):
assert out_file.endswith('.dscalar.nii')
reader = CiftiReader(data_file)
data = reader.get_data()
else:
raise ValueError(1)
# calculate
ages = np.array(pd.read_csv(info_file)['age in years'])
coefs = np.polyfit(ages, data, deg)
n_row = coefs.shape[0]
if deg == 1:
assert n_row == 2
row_names = ['coef', 'intercept']
else:
row_names = [''] * n_row
raise Warning("Can't provide row names for degree:", deg)
# save
if out_file.endswith('.csv'):
df = pd.DataFrame(coefs, row_names, reader.columns)
df.to_csv(out_file)
else:
save2cifti(out_file, coefs, reader.brain_models(), row_names)
def merge_data(dataset_name, meas_name):
"""
把所有被试的数据合并到一个cifti文件里
Args:
dataset_name (str): HCPD | HCPA
meas_name (str): thickness | myelin
"""
# outputs
out_file = pjoin(work_dir, f'{dataset_name}_{meas_name}.dscalar.nii')
# prepare
dataset_dir = dataset_name2dir[dataset_name]
meas2file = {
'myelin':
pjoin(
dataset_dir,
'fmriresults01/{sid}_V1_MR/MNINonLinear/fsaverage_LR32k/'
'{sid}_V1_MR.MyelinMap_BC_MSMAll.32k_fs_LR.dscalar.nii'),
'thickness':
pjoin(
dataset_dir,
'fmriresults01/{sid}_V1_MR/MNINonLinear/fsaverage_LR32k/'
'{sid}_V1_MR.thickness_MSMAll.32k_fs_LR.dscalar.nii')
}
df = pd.read_csv(dataset_name2info[dataset_name])
n_subj = df.shape[0]
data = np.zeros((n_subj, LR_count_32k), np.float64)
# calculate
for subj_idx, subj_id in enumerate(df['subID']):
time1 = time.time()
meas_file = meas2file[meas_name].format(sid=subj_id)
data[subj_idx] = nib.load(meas_file).get_fdata()[0]
print(f'Finished: {subj_idx+1}/{n_subj},'
f'cost: {time.time() - time1} seconds.')
# save
mmp_reader = CiftiReader(mmp_map_file)
save2cifti(out_file, data, mmp_reader.brain_models(), df['subID'])
def separate_networks():
"""
把ColeNetwork的12个网络分到单独的map里。
每个map的MMP parcel的label保留原来的样子。
需要注意的是multimodal_glasser的MMP_mpmLR32k.dlabel.nii中，
ROI label的编号1~180是右脑，181~360是左脑。0对应的是???
而cortex_parcel_network_assignments.mat中0~359的index是左脑在先
"""
# outputs
out_file = pjoin(work_dir, 'networks.dlabel.nii')
# load
mmp_reader = CiftiReader(mmp_map_file)
mmp_map = mmp_reader.get_data()[0]
lbl_tab_raw = mmp_reader.label_tables()[0]
roi2net = loadmat(cole_net_assignment_file)['netassignments'][:, 0]
roi2net = np.r_[roi2net[180:], roi2net[:180]]
net_labels = np.unique(roi2net)
# prepare
data = np.zeros((len(net_labels), len(mmp_map)), dtype=np.uint16)
map_names = []
label_tables = []
# calculate
for net_idx, net_lbl in enumerate(net_labels):
roi_labels = np.where(roi2net == net_lbl)[0] + 1
lbl_tab = nib.cifti2.cifti2.Cifti2LabelTable()
lbl_tab[0] = lbl_tab_raw[0]
for roi_lbl in roi_labels:
data[net_idx, mmp_map == roi_lbl] = roi_lbl
lbl_tab[roi_lbl] = lbl_tab_raw[roi_lbl]
map_names.append(cole_label2name[net_lbl])
label_tables.append(lbl_tab)
# save
save2cifti(out_file,
data,
mmp_reader.brain_models(),
map_names,
label_tables=label_tables)
def calc_mean_meas_map(meas='thickness'):
"""
得到1080个被试的平均map
"""
import numpy as np
from magicbox.io.io import CiftiReader, save2nifti
from cxy_hcp_ffa.lib.predefine import hemi2stru
# inputs
hemis = ('lh', 'rh')
subj_file = pjoin(proj_dir, 'analysis/s2/subject_id')
meas_id_file = pjoin(proj_dir, 'data/HCP/subject_id_1096')
meas2file = {
'thickness':
'/nfs/p1/public_dataset/datasets/hcp/DATA/'
'HCP_S1200_GroupAvg_v1/HCP_S1200_GroupAvg_v1/'
'S1200.All.thickness_MSMAll.32k_fs_LR.dscalar.nii',
'myelin':
'/nfs/p1/public_dataset/datasets/hcp/DATA/'
'HCP_S1200_GroupAvg_v1/HCP_S1200_GroupAvg_v1/'
'S1200.All.MyelinMap_BC_MSMAll.32k_fs_LR.dscalar.nii',
'va':
'/nfs/p1/public_dataset/datasets/hcp/DATA/'
'HCP_S1200_GroupAvg_v1/HCP_S1200_GroupAvg_v1/'
'S1200.All.midthickness_MSMAll_va.32k_fs_LR.dscalar.nii'
}
# outputs
out_file = pjoin(work_dir, '{meas}_mean-1080_{hemi}.nii.gz')
# prepare
subj_ids = open(subj_file).read().splitlines()
meas_reader = CiftiReader(meas2file[meas])
meas_ids = open(meas_id_file).read().splitlines()
meas_indices = [meas_ids.index(i) for i in subj_ids]
# calculate
for hemi in hemis:
meas_maps = meas_reader.get_data(hemi2stru[hemi], True)[meas_indices]
meas_mean = np.mean(meas_maps, 0)
meas_mean = np.expand_dims(meas_mean, (1, 2))
save2nifti(out_file.format(meas=meas, hemi=hemi), meas_mean)
def get_curvature(hemi='lh'):
"""
把45个retest被试在retest session的curvature合并成左右脑两个nifti文件，
主要是为了我那个程序在标定个体ROI的时候读取和显示曲率。
之前服务器上没有retest的结构数据，我想当然地认为同一个被试的沟回曲率在
两次测量应该是一模一样的，所以在标定v1和v2版ROI的时候参照的是test session的曲率；
这次我下了retest的结构数据，决定用retest session的曲率校对一下retest个体ROI。
"""
import numpy as np
from magicbox.io.io import CiftiReader, save2nifti
from cxy_hcp_ffa.lib.predefine import hemi2stru
# inputs
hemis = ('lh', 'rh')
fpath = '/nfs/m1/hcp/retest/{0}/MNINonLinear/fsaverage_LR32k/'\
'{0}.curvature_MSMAll.32k_fs_LR.dscalar.nii'
subj_id_file = pjoin(work_dir, 'subject_id')
# outputs
out_file = pjoin(work_dir, 'curvature_{hemi}.nii.gz')
# prepare
subj_ids = open(subj_id_file).read().splitlines()
n_subj = len(subj_ids)
hemi2data = {}
for hemi in hemis:
hemi2data[hemi] = np.zeros((32492, 1, 1, n_subj), np.float64)
# calculate
for subj_idx, subj_id in enumerate(subj_ids):
reader = CiftiReader(fpath.format(subj_id))
for hemi in hemis:
data_tmp = reader.get_data(hemi2stru[hemi], True)[0]
hemi2data[hemi][:, 0, 0, subj_idx] = data_tmp
print(f'Finished: {subj_idx+1}/{n_subj}')
# save
for hemi in hemis:
save2nifti(out_file.format(hemi=hemi), hemi2data[hemi])
def mask_maps(data_file, atlas_name, roi_name, out_file):
"""
把data map在指定atlas的ROI以外的部分全赋值为nan
Args:
data_file (str): end with .dscalar.nii
shape=(n_map, LR_count_32k)
atlas_name (str):
roi_name (str):
out_file (str):
"""
# prepare
reader = CiftiReader(data_file)
data = reader.get_data()
atlas = Atlas(atlas_name)
assert atlas.maps.shape == (1, LR_count_32k)
roi_idx_map = atlas.maps[0] == atlas.roi2label[roi_name]
# calculate
data[:, ~roi_idx_map] = np.nan
# save
save2cifti(out_file, data, reader.brain_models(), reader.map_names())
def get_valid_id(sess=1, run='AP'):
import os
import time
import pandas as pd
from magicbox.io.io import CiftiReader
# inputs
subj_info_file = pjoin(dev_dir, 'HCPD_SubjInfo.csv')
# outputs
log_file = pjoin(work_dir, f'get_valid_id_log_{sess}_{run}')
out_file = pjoin(work_dir, f'rfMRI_REST{sess}_{run}_id')
subj_info = pd.read_csv(subj_info_file)
subj_ids = subj_info['subID'].to_list()
n_subj = len(subj_ids)
maps_files = '/nfs/e1/HCPD/fmriresults01/{0}/MNINonLinear/Results/rfMRI_REST{1}_{2}/rfMRI_REST{1}_{2}_Atlas_MSMAll_hp0_clean.dtseries.nii'
valid_ids = []
log_writer = open(log_file, 'w')
for idx, subj_id in enumerate(subj_ids, 1):
time1 = time.time()
maps_file = maps_files.format(f'{subj_id}_V1_MR', sess, run)
if not os.path.exists(maps_file):
msg = f'{maps_file} is not exist.'
print(msg)
log_writer.write(f'{msg}\n')
continue
try:
data = CiftiReader(maps_file).get_data()
except OSError:
msg = f'{maps_file} meets OSError.'
print(msg)
log_writer.write(f'{msg}\n')
continue
if data.shape[0] != 478:
msg = f'The number of time points in {maps_file} is not 478.'
print(msg)
log_writer.write(f'{msg}\n')
continue
valid_ids.append(subj_id)
print(
f'Finished: {idx}/{n_subj}, cost: {time.time() - time1} seconds.')
log_writer.close()
# save out
with open(out_file, 'w') as wf:
wf.write('\n'.join(valid_ids))
def kendall2cifti(data_file, rois, atlas_name, out_file):
"""
把指定ROI的Tau值和p值存成cifti格式 方便可视化在大脑上
"""
# prepare
df = pd.read_csv(data_file, index_col=0)
atlas = Atlas(atlas_name)
assert atlas.maps.shape == (1, LR_count_32k)
out_data = np.ones((2, LR_count_32k)) * np.nan
# calculate
for roi in rois:
mask = atlas.maps[0] == atlas.roi2label[roi]
out_data[0, mask] = df.loc['tau', roi]
out_data[1, mask] = -np.log10(df.loc['p', roi])
# save
save2cifti(out_file, out_data,
CiftiReader(mmp_map_file).brain_models(), ['tau', '-lg(p)'])
def calc_meas(meas_name='activ', atlas_name='MPM'):
import time
import numpy as np
import pickle as pkl
import nibabel as nib
from magicbox.io.io import CiftiReader
from cxy_hcp_ffa.lib.predefine import roi2label, hemi2stru
# inputs
hemis = ('lh', 'rh')
rois = ('pFus-face', 'mFus-face')
subj_file_45 = pjoin(proj_dir, 'data/HCP/wm/analysis_s2/'
'retest/subject_id')
subj_file_1080 = pjoin(proj_dir, 'analysis/s2/subject_id')
subj_file_1096 = pjoin(proj_dir, 'data/HCP/subject_id_1096')
meas2file_test = {
'thickness':
'/nfs/p1/public_dataset/datasets/hcp/DATA/'
'HCP_S1200_GroupAvg_v1/HCP_S1200_GroupAvg_v1/'
'S1200.All.thickness_MSMAll.32k_fs_LR.dscalar.nii',
'myelin':
'/nfs/p1/public_dataset/datasets/hcp/DATA/'
'HCP_S1200_GroupAvg_v1/HCP_S1200_GroupAvg_v1/'
'S1200.All.MyelinMap_BC_MSMAll.32k_fs_LR.dscalar.nii',
'activ':
pjoin(proj_dir, 'analysis/s2/activation.dscalar.nii')
}
retest_dir = '/nfs/m1/hcp/retest'
meas2file_retest = {
'thickness':
pjoin(
retest_dir, '{0}/MNINonLinear/fsaverage_LR32k/'
'{0}.thickness_MSMAll.32k_fs_LR.dscalar.nii'),
'myelin':
pjoin(
retest_dir, '{0}/MNINonLinear/fsaverage_LR32k/'
'{0}.MyelinMap_BC_MSMAll.32k_fs_LR.dscalar.nii'),
'activ':
pjoin(
retest_dir, '{0}/MNINonLinear/Results/tfMRI_WM/'
'tfMRI_WM_hp200_s2_level2_MSMAll.feat/GrayordinatesStats/'
'cope20.feat/zstat1.dtseries.nii')
}
atlas2file = {
'MPM':
pjoin(
proj_dir, 'analysis/s2/1080_fROI/refined_with_Kevin/'
'MPM_v3_{hemi}_0.25.nii.gz'),
'ROIv3':
pjoin(
proj_dir, 'analysis/s2/1080_fROI/refined_with_Kevin/'
'rois_v3_{hemi}.nii.gz')
}
# outputs
out_file = pjoin(work_dir, f'{meas_name}_{atlas_name}.pkl')
# prepare subject IDs
subj_ids_45 = open(subj_file_45).read().splitlines()
subj_ids_1080 = open(subj_file_1080).read().splitlines()
subj_ids_1096 = open(subj_file_1096).read().splitlines()
retest_idx_in_1080 = [subj_ids_1080.index(i) for i in subj_ids_45]
retest_idx_in_1096 = [subj_ids_1096.index(i) for i in subj_ids_45]
n_subj = len(subj_ids_45)
# prepare atlas
hemi2atlas = {}
for hemi in hemis:
atlas_tmp = nib.load(atlas2file[atlas_name].format(hemi=hemi))
atlas_tmp = atlas_tmp.get_fdata().squeeze()
if atlas_name == 'MPM':
atlas_tmp = np.expand_dims(atlas_tmp, 0)
hemi2atlas[hemi] = np.repeat(atlas_tmp, n_subj, 0)
elif atlas_name == 'ROIv3':
hemi2atlas[hemi] = atlas_tmp.T[retest_idx_in_1080]
else:
raise ValueError('Not supported atlas name:', atlas_name)
# prepare test measurement maps
if meas_name == 'activ':
retest_idx_vec = retest_idx_in_1080
else:
retest_idx_vec = retest_idx_in_1096
meas_reader_test = CiftiReader(meas2file_test[meas_name])
hemi2maps_test = {}
for hemi in hemis:
hemi2maps_test[hemi] = meas_reader_test.get_data(
hemi2stru[hemi], True)[retest_idx_vec]
# prepare out data
data = {}
for hemi in hemis:
for roi in rois:
roi_name = roi.split('-')[0]
data[f"{hemi}_{roi_name}_test"] = np.ones(n_subj) * np.nan
data[f"{hemi}_{roi_name}_retest"] = np.ones(n_subj) * np.nan
# calculate
for subj_idx, subj_id in enumerate(subj_ids_45):
time1 = time.time()
meas_reader_retest = CiftiReader(
meas2file_retest[meas_name].format(subj_id))
for hemi in hemis:
meas_map_test = hemi2maps_test[hemi][subj_idx]
meas_map_retest = meas_reader_retest.get_data(
hemi2stru[hemi], True)[0]
atlas_map = hemi2atlas[hemi][subj_idx]
for roi in rois:
idx_vec = atlas_map == roi2label[roi]
if np.any(idx_vec):
roi_name = roi.split('-')[0]
data[f"{hemi}_{roi_name}_test"][subj_idx] =\
np.mean(meas_map_test[idx_vec])
data[f"{hemi}_{roi_name}_retest"][subj_idx] =\
np.mean(meas_map_retest[idx_vec])
print(
f'Finished {subj_idx+1}/{n_subj}: cost {time.time()-time1} seconds'
)
# save
pkl.dump(data, open(out_file, 'wb'))
def calc_gdist(method='peak'):
"""Calculate geodesic distance between each two ROIs.
Args:
method (str, optional): 'peak' or 'min'
If 'peak', use the distance between two vertices
with peak activation values in two ROIs respectively.
If 'min', use the minimum distance of pair-wise
vertices between the two ROIs.
Defaults to 'peak'.
"""
import os
import time
import gdist
import numpy as np
import pandas as pd
import nibabel as nib
from cxy_hcp_ffa.lib.predefine import roi2label, hemi2stru
from magicbox.io.io import CiftiReader
# inputs
rois = ('IOG-face', 'pFus-face', 'mFus-face')
hemis = ('lh', 'rh')
hemi2Hemi = {'lh': 'L', 'rh': 'R'}
subj_file = pjoin(proj_dir, 'analysis/s2/subject_id')
roi_file = pjoin(work_dir, 'rois_v3_{}.nii.gz')
geo_file = '/nfs/m1/hcp/{sid}/T1w/fsaverage_LR32k/' \
'{sid}.{Hemi}.midthickness_MSMAll.32k_fs_LR.surf.gii'
activ_file = pjoin(proj_dir, 'analysis/s2/activation.dscalar.nii')
# outputs
log_file = pjoin(work_dir, f'gdist_{method}_log')
out_file = pjoin(work_dir, f'gdist_{method}.csv')
# preparation
subj_ids = open(subj_file).read().splitlines()
n_subj = len(subj_ids)
activ_reader = CiftiReader(activ_file)
out_dict = {}
for hemi in hemis:
for roi1_idx, roi1 in enumerate(rois[:-1]):
for roi2 in rois[roi1_idx + 1:]:
k = f"{hemi}_{roi1.split('-')[0]}-{roi2.split('-')[0]}"
out_dict[k] = np.ones(n_subj, dtype=np.float64) * np.nan
log_lines = []
# calculation
for hemi in hemis:
roi_maps = nib.load(roi_file.format(hemi)).get_fdata().squeeze().T
activ_maps = activ_reader.get_data(hemi2stru[hemi], True)
assert roi_maps.shape == activ_maps.shape
for subj_idx, subj_id in enumerate(subj_ids):
time1 = time.time()
roi_map = roi_maps[subj_idx]
activ_map = activ_maps[subj_idx]
g_file = geo_file.format(sid=subj_id, Hemi=hemi2Hemi[hemi])
if not os.path.exists(g_file):
log_lines.append(f'{g_file} does not exist.')
continue
geo = nib.load(g_file)
coords = geo.get_arrays_from_intent('NIFTI_INTENT_POINTSET')[0]
coords = coords.data.astype(np.float64)
faces = geo.get_arrays_from_intent('NIFTI_INTENT_TRIANGLE')[0]
faces = faces.data.astype(np.int32)
for roi1_idx, roi1 in enumerate(rois[:-1]):
roi1_idx_map = roi_map == roi2label[roi1]
if np.any(roi1_idx_map):
for roi2 in rois[roi1_idx + 1:]:
roi2_idx_map = roi_map == roi2label[roi2]
if np.any(roi2_idx_map):
k = f"{hemi}_{roi1.split('-')[0]}-"\
f"{roi2.split('-')[0]}"
if method == 'peak':
roi1_max = np.max(activ_map[roi1_idx_map])
roi2_max = np.max(activ_map[roi2_idx_map])
roi1_idx_map =\
np.logical_and(roi1_idx_map,
activ_map == roi1_max)
roi2_idx_map =\
np.logical_and(roi2_idx_map,
activ_map == roi2_max)
roi1_vertices = np.where(roi1_idx_map)[0]
roi1_vertices = roi1_vertices.astype(np.int32)
n_vtx1 = len(roi1_vertices)
roi2_vertices = np.where(roi2_idx_map)[0]
roi2_vertices = roi2_vertices.astype(np.int32)
n_vtx2 = len(roi2_vertices)
if n_vtx1 > 1 or n_vtx2 > 1:
msg = f'{subj_id}: {roi1} vs {roi2} '\
f'has multiple peaks.'
log_lines.append(msg)
ds = []
for src_vtx in roi1_vertices:
src_vtx = np.array([src_vtx], np.int32)
ds_tmp = \
gdist.compute_gdist(coords, faces,
src_vtx,
roi2_vertices)
ds.extend(ds_tmp)
out_dict[k][subj_idx] = np.mean(ds)
elif n_vtx1 == 1 and n_vtx2 == 1:
ds = gdist.compute_gdist(
coords, faces, roi1_vertices,
roi2_vertices)
assert len(ds) == 1
out_dict[k][subj_idx] = ds[0]
else:
raise RuntimeError("Impossible!")
elif method == 'min':
roi1_vertices = np.where(roi1_idx_map)[0]
roi1_vertices = roi1_vertices.astype(np.int32)
roi2_vertices = np.where(roi2_idx_map)[0]
roi2_vertices = roi2_vertices.astype(np.int32)
ds = gdist.compute_gdist(
coords, faces, roi1_vertices,
roi2_vertices)
out_dict[k][subj_idx] = np.min(ds)
else:
raise ValueError(f'Not supported method: '
f'{method}')
print(f'Finished: {subj_idx+1}/{n_subj}, '
f'cost {time.time()-time1} seconds.')
# save out
out_df = pd.DataFrame(out_dict)
out_df.to_csv(out_file, index=False)
out_log = '\n'.join(log_lines)
open(log_file, 'w').write(out_log)
def test_retest_icc(hemi='lh'):
import numpy as np
import pickle as pkl
import nibabel as nib
from cxy_hcp_ffa.lib.predefine import roi2label, hemi2stru
from cxy_hcp_ffa.lib.heritability import icc
from magicbox.io.io import CiftiReader
# inputs
roi_names = ('pFus-face', 'mFus-face')
test_file = pjoin(proj_dir, 'analysis/s2/activation.dscalar.nii')
retest_file = pjoin(work_dir, 'activation.dscalar.nii')
subj_test_file = pjoin(proj_dir, 'analysis/s2/subject_id')
subj_retest_file = pjoin(work_dir, 'subject_id')
mpm_file = pjoin(
proj_dir, 'analysis/s2/1080_fROI/refined_with_Kevin/'
f'MPM_v3_{hemi}_0.25_FFA.nii.gz')
individual_file = pjoin(
proj_dir, 'analysis/s2/1080_fROI/refined_with_Kevin/'
f'rois_v3_{hemi}.nii.gz')
# outputs
out_file = pjoin(work_dir, f'ICC_{hemi}.pkl')
# prepare
subj_ids_test = open(subj_test_file).read().splitlines()
subj_ids_retest = open(subj_retest_file).read().splitlines()
retest_indices = [subj_ids_test.index(i) for i in subj_ids_retest]
test_maps = CiftiReader(test_file).get_data(hemi2stru[hemi],
True)[retest_indices]
retest_maps = CiftiReader(retest_file).get_data(hemi2stru[hemi], True)
mpm_mask = nib.load(mpm_file).get_fdata().squeeze()
individual_mask = nib.load(
individual_file).get_fdata().squeeze().T[retest_indices]
# calculate
n_roi = len(roi_names)
mpm_iccs = np.ones((3, n_roi)) * np.nan
individual_iccs = np.ones((3, n_roi)) * np.nan
for i, roi_name in enumerate(roi_names):
label = roi2label[roi_name]
mpm_roi_idx_arr = mpm_mask == label
individual_roi_idx_arr = individual_mask == label
valid_subj_idx_vec = np.any(individual_roi_idx_arr, 1)
individual_roi_idx_arr = individual_roi_idx_arr[valid_subj_idx_vec]
test_maps_tmp = test_maps[valid_subj_idx_vec]
retest_maps_tmp = retest_maps[valid_subj_idx_vec]
n_subj_valid = np.sum(valid_subj_idx_vec)
print(f'{hemi}_{roi_name}: {n_subj_valid} valid subjects')
# calculate ICC for MPM
test_series_mpm = np.mean(test_maps_tmp[:, mpm_roi_idx_arr], 1)
retest_series_mpm = np.mean(retest_maps_tmp[:, mpm_roi_idx_arr], 1)
mpm_iccs[:, i] = icc(test_series_mpm, retest_series_mpm, 10000, 95)
# calculate ICC for individual
test_series_ind = np.zeros((n_subj_valid, ))
retest_series_ind = np.zeros((n_subj_valid, ))
for j in range(n_subj_valid):
individual_roi_idx_vec = individual_roi_idx_arr[j]
test_series_ind[j] = np.mean(
test_maps_tmp[j][individual_roi_idx_vec])
retest_series_ind[j] = np.mean(
retest_maps_tmp[j][individual_roi_idx_vec])
individual_iccs[:, i] = icc(test_series_ind, retest_series_ind, 10000,
95)
# save
out_data = {
'roi_name': roi_names,
'mpm': mpm_iccs,
'individual': individual_iccs
}
pkl.dump(out_data, open(out_file, 'wb'))
def prepare_series(sess=1, run='AP'):
import numpy as np
import nibabel as nib
import pickle as pkl
from cxy_hcp_ffa.lib.predefine import hemi2stru, roi2label
from magicbox.io.io import CiftiReader
# prepare seeds
seeds = ('IOG-face', 'pFus-face', 'mFus-face')
n_seed = len(seeds)
seed_mask_lh_file = pjoin(
proj_dir, 'analysis/s2/1080_fROI/'
'refined_with_Kevin/MPM_v3_lh_0.25.nii.gz')
seed_mask_rh_file = pjoin(
proj_dir, 'analysis/s2/1080_fROI/'
'refined_with_Kevin/MPM_v3_rh_0.25.nii.gz')
hemi2seed_mask = {
'lh': nib.load(seed_mask_lh_file).get_fdata().squeeze(),
'rh': nib.load(seed_mask_rh_file).get_fdata().squeeze()
}
# prepare targets
trg_file = '/nfs/p1/atlases/multimodal_glasser/surface/'\
'MMP_mpmLR32k.dlabel.nii'
trg_reader = CiftiReader(trg_file)
hemi2trg_mask = {
'lh': trg_reader.get_data(hemi2stru['lh'], True).squeeze(),
'rh': trg_reader.get_data(hemi2stru['rh'], True).squeeze()
}
trg_labels = np.unique(np.r_[hemi2trg_mask['lh'], hemi2trg_mask['rh']])
trg_labels = trg_labels[trg_labels != 0].astype(int).tolist()
n_trg = len(trg_labels)
# prepare series dictionary
subj_id_file = pjoin(work_dir, 'rfMRI_REST_4run_id')
subj_ids = open(subj_id_file).read().splitlines()
n_subj = len(subj_ids)
n_tp = 478 # the number of time points
seed_lh_dict = {
'shape': 'n_subject x n_seed x n_time_point',
'subject': subj_ids,
'seed': seeds,
'rfMRI': np.ones((n_subj, n_seed, n_tp)) * np.nan
}
seed_rh_dict = {
'shape': 'n_subject x n_seed x n_time_point',
'subject': subj_ids,
'seed': seeds,
'rfMRI': np.ones((n_subj, n_seed, n_tp)) * np.nan
}
hemi2seed_dict = {'lh': seed_lh_dict, 'rh': seed_rh_dict}
trg_dict = {
'shape': 'n_subject x n_target x n_time_point',
'subject': subj_ids,
'trg_label': trg_labels,
'rfMRI': np.ones((n_subj, n_trg, n_tp)) * np.nan
}
global_dict = {
'shape': 'n_subject x n_global x n_time_point',
'subject': subj_ids,
'global': ['lr', 'lh', 'rh'],
'rfMRI': np.ones((n_subj, 3, n_tp)) * np.nan
}
# prepare outputs
out_seed_lh = pjoin(work_dir, f'rfMRI{sess}_{run}_MPM_lh.pkl')
out_seed_rh = pjoin(work_dir, f'rfMRI{sess}_{run}_MPM_rh.pkl')
out_trg = pjoin(work_dir, f'rfMRI{sess}_{run}_MMP.pkl')
out_global = pjoin(work_dir, f'rfMRI{sess}_{run}_global.pkl')
# start
maps_files = '/nfs/e1/HCPD/fmriresults01/{0}/MNINonLinear/'\
'Results/rfMRI_REST{1}_{2}/'\
'rfMRI_REST{1}_{2}_Atlas_MSMAll_hp0_clean.dtseries.nii'
for subj_idx, subj_id in enumerate(subj_ids):
print('Progress: {}/{}'.format(subj_idx + 1, n_subj))
# prepare maps
maps_file = maps_files.format(f'{subj_id}_V1_MR', sess, run)
maps_reader = CiftiReader(maps_file)
hemi2maps = {
'lh': maps_reader.get_data(hemi2stru['lh'], True),
'rh': maps_reader.get_data(hemi2stru['rh'], True)
}
for hemi in ['lh', 'rh']:
maps = hemi2maps[hemi]
# seed dict
seed_mask = hemi2seed_mask[hemi]
for seed_idx, seed in enumerate(seeds):
lbl = roi2label[seed]
hemi2seed_dict[hemi]['rfMRI'][subj_idx, seed_idx] =\
np.mean(maps[:, seed_mask == lbl], 1)
# target dict
trg_mask = hemi2trg_mask[hemi]
tmp_labels = np.unique(trg_mask)
tmp_labels = tmp_labels[tmp_labels != 0].astype(int)
for lbl in tmp_labels:
trg_idx = trg_labels.index(lbl)
trg_dict['rfMRI'][subj_idx, trg_idx] =\
np.mean(maps[:, trg_mask == lbl], 1)
# global dict
global_idx = global_dict['global'].index(hemi)
global_dict['rfMRI'][subj_idx, global_idx] = np.mean(maps, 1)
maps = np.c_[hemi2maps['lh'], hemi2maps['rh']]
global_idx = global_dict['global'].index('lr')
global_dict['rfMRI'][subj_idx, global_idx] = np.mean(maps, 1)
del maps
pkl.dump(hemi2seed_dict['lh'], open(out_seed_lh, 'wb'))
pkl.dump(hemi2seed_dict['rh'], open(out_seed_rh, 'wb'))
pkl.dump(trg_dict, open(out_trg, 'wb'))
pkl.dump(global_dict, open(out_global, 'wb'))
def prepare_series_ind(sess=1, run='LR'):
import numpy as np
import nibabel as nib
import pickle as pkl
from cxy_hcp_ffa.lib.predefine import hemi2stru, roi2label
from magicbox.io.io import CiftiReader
# prepare subjects info
subj_id_file = pjoin(work_dir, 'rfMRI_REST_id')
subj_file_1080 = pjoin(proj_dir, 'analysis/s2/subject_id')
subj_ids = open(subj_id_file).read().splitlines()
subj_ids_1080 = open(subj_file_1080).read().splitlines()
subj_idx_in_1080 = [subj_ids_1080.index(i) for i in subj_ids]
n_subj = len(subj_ids)
# prepare seeds
seeds = ('IOG-face', 'pFus-face', 'mFus-face')
n_seed = len(seeds)
seed_mask_lh_file = pjoin(
proj_dir, 'analysis/s2/1080_fROI/'
'refined_with_Kevin/rois_v3_lh.nii.gz')
seed_mask_rh_file = pjoin(
proj_dir, 'analysis/s2/1080_fROI/'
'refined_with_Kevin/rois_v3_rh.nii.gz')
hemi2seed_mask = {
'lh':
nib.load(seed_mask_lh_file).get_fdata().squeeze().T[subj_idx_in_1080],
'rh':
nib.load(seed_mask_rh_file).get_fdata().squeeze().T[subj_idx_in_1080]
}
# prepare series dictionary
n_tp = 1200 # the number of time points
seed_lh_dict = {
'shape': 'n_subject x n_seed x n_time_point',
'subject': subj_ids,
'seed': seeds,
'rfMRI': np.ones((n_subj, n_seed, n_tp)) * np.nan
}
seed_rh_dict = {
'shape': 'n_subject x n_seed x n_time_point',
'subject': subj_ids,
'seed': seeds,
'rfMRI': np.ones((n_subj, n_seed, n_tp)) * np.nan
}
hemi2seed_dict = {'lh': seed_lh_dict, 'rh': seed_rh_dict}
# prepare outputs
out_seed_lh = pjoin(work_dir, f'rfMRI{sess}_{run}_ROIv3_lh.pkl')
out_seed_rh = pjoin(work_dir, f'rfMRI{sess}_{run}_ROIv3_rh.pkl')
# start
maps_files = '/nfs/m1/hcp/retest/{0}/MNINonLinear/Results/rfMRI_REST{1}_{2}/'\
'rfMRI_REST{1}_{2}_Atlas_MSMAll_hp2000_clean.dtseries.nii'
for subj_idx, subj_id in enumerate(subj_ids):
print('Progress: {}/{}'.format(subj_idx + 1, n_subj))
# prepare maps
maps_file = maps_files.format(subj_id, sess, run)
maps_reader = CiftiReader(maps_file)
hemi2maps = {
'lh': maps_reader.get_data(hemi2stru['lh'], True),
'rh': maps_reader.get_data(hemi2stru['rh'], True)
}
for hemi in ['lh', 'rh']:
maps = hemi2maps[hemi]
# seed dict
seed_mask = hemi2seed_mask[hemi]
for seed_idx, seed in enumerate(seeds):
idx_vec = seed_mask[subj_idx] == roi2label[seed]
if np.any(idx_vec):
hemi2seed_dict[hemi]['rfMRI'][subj_idx, seed_idx] =\
np.mean(maps[:, idx_vec], 1)
# save
pkl.dump(hemi2seed_dict['lh'], open(out_seed_lh, 'wb'))
pkl.dump(hemi2seed_dict['rh'], open(out_seed_rh, 'wb'))
def calc_meas_retest(roi_type):
"""
用45个被试在test或retest中定出的个体FFA去提取各自在retest的face selectivity。
Args:
roi_type (str): ROI-test or ROI-retest
"""
import time
import numpy as np
import pandas as pd
import nibabel as nib
from magicbox.io.io import CiftiReader
from cxy_hcp_ffa.lib.predefine import hemi2stru, roi2label
# inputs
hemis = ('lh', 'rh')
rois = ('pFus-face', 'mFus-face')
subj_file_45 = pjoin(proj_dir, 'data/HCP/wm/analysis_s2/'
'retest/subject_id')
subj_file_1080 = pjoin(proj_dir, 'analysis/s2/subject_id')
meas_file = '/nfs/m1/hcp/retest/{0}/MNINonLinear/Results/tfMRI_WM/'\
'tfMRI_WM_hp200_s2_level2_MSMAll.feat/GrayordinatesStats/'\
'cope20.feat/zstat1.dtseries.nii'
roi_type2file = {
'ROI-test':
pjoin(
proj_dir, 'analysis/s2/1080_fROI/refined_with_Kevin/'
'rois_v3_{hemi}.nii.gz'),
'ROI-retest':
pjoin(
proj_dir, 'analysis/s2/1080_fROI/refined_with_Kevin/'
'retest/rois_{hemi}_v2.nii.gz')
}
roi_file = roi_type2file[roi_type]
# outputs
out_file = pjoin(work_dir, f'activ-retest_{roi_type}.csv')
# prepare
subj_ids_45 = open(subj_file_45).read().splitlines()
subj_ids_1080 = open(subj_file_1080).read().splitlines()
retest_idx_in_1080 = [subj_ids_1080.index(i) for i in subj_ids_45]
n_subj = len(subj_ids_45)
out_df = pd.DataFrame(index=range(n_subj))
# prepare atlas
hemi2atlas = {}
for hemi in hemis:
atlas = nib.load(roi_file.format(hemi=hemi)).get_fdata().squeeze().T
if roi_type == 'ROI-test':
hemi2atlas[hemi] = atlas[retest_idx_in_1080]
else:
hemi2atlas[hemi] = atlas
# calculate
for idx, subj_id in enumerate(subj_ids_45):
time1 = time.time()
meas_reader = CiftiReader(meas_file.format(subj_id))
for hemi in hemis:
meas_map = meas_reader.get_data(hemi2stru[hemi], True)[0]
roi_map = hemi2atlas[hemi][idx]
for roi in rois:
col = f"{hemi}_{roi.split('-')[0]}"
roi_idx_map = roi_map == roi2label[roi]
if np.any(roi_idx_map):
out_df.loc[idx, col] = np.mean(meas_map[roi_idx_map])
print(f'Finished {idx+1}/{n_subj}: cost {time.time()-time1} seconds')
# save
out_df.to_csv(out_file, index=False)