diff --git a/notebooks/demo.ipynb b/notebooks/demo.ipynb
index 528dde4918f43de23e65de008786e3abe21d0d0e..383798919e142896b743c8bcf9398af301abf3aa 100644
--- a/notebooks/demo.ipynb
+++ b/notebooks/demo.ipynb
@@ -10,43 +10,12 @@
},
{
"cell_type": "code",
- "execution_count": 1,
+ "execution_count": null,
"id": "5b441713-77b4-4e57-a188-9959c09784bd",
"metadata": {
- "execution": {
- "iopub.execute_input": "2022-02-11T10:40:43.234120Z",
- "iopub.status.busy": "2022-02-11T10:40:43.233485Z",
- "iopub.status.idle": "2022-02-11T10:40:44.641758Z",
- "shell.execute_reply": "2022-02-11T10:40:44.641057Z",
- "shell.execute_reply.started": "2022-02-11T10:40:43.234047Z"
- }
+ "tags": []
},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "2022-02-11 11:40:43,461 - WARNING - tvb.simulator.common - psutil module not available: no warnings will be issued when a\n",
- " simulation may require more memory than available\n",
- " INFO log level set to INFO\n"
- ]
- },
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "/home/izaak/local_repos/nostromo/tvb-ebrains-data/env/lib/python3.6/site-packages/tvb/datatypes/surfaces.py:63: UserWarning: Geodesic distance module is unavailable; some functionality for surfaces will be unavailable.\n",
- " warnings.warn(msg)\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "WARNING Token required outside collaboratory environment. Set EBRAINS_TOKEN in the environment or provide directly.\n"
- ]
- }
- ],
+ "outputs": [],
"source": [
"from tvb_ebrains_data import Brains1000Dataset"
]
@@ -374,11 +343,136 @@
"source": [
"W.shape, D.shape"
]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "id": "896ad8f3-c824-48d7-a2dc-8503a808294e",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "from tvb_ebrains_data import HCPDomhof"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "4e730e39-68cf-4abc-b0dd-14abd74e82e3",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [],
+ "source": [
+ "dset = HCPDomhof(data_root='.')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "id": "4e58c1d7-743d-41aa-9a43-bf7b90cc195d",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "['093', '124', '106', '020', '077']"
+ ]
+ },
+ "execution_count": 6,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "dset.subjects[:5]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "id": "5dcd7edd-8067-4d37-a55b-46a831279b75",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "(1200, 200)"
+ ]
+ },
+ "execution_count": 7,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "dset.load_bold('093').shape"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "id": "a1a6c4ef-5a0f-4f81-a952-c0117c168e85",
+ "metadata": {
+ "tags": []
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Help on method load_sc in module tvb_ebrains_data.data:\n",
+ "\n",
+ "load_sc(subj) method of tvb_ebrains_data.data.HCPDomhof instance\n",
+ " Load structural connectivity.\n",
+ " \n",
+ " Parameters\n",
+ " ----------\n",
+ " subj : str\n",
+ " Subject id from `self.subjects`.\n",
+ " \n",
+ " Returns\n",
+ " -------\n",
+ " weights: ndarray\n",
+ " Weights matrix [N,N].\n",
+ " tract_length: ndarray\n",
+ " Tract length matrix [N,N].\n",
+ "\n"
+ ]
+ }
+ ],
+ "source": [
+ "help(dset.load_sc)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "id": "6997cabe-3541-4ea5-8dff-cf635c853473",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "(200, 200)"
+ ]
+ },
+ "execution_count": 10,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "dset.load_sc('093')[0].shape"
+ ]
}
],
"metadata": {
"kernelspec": {
- "display_name": "Python 3",
+ "display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
@@ -392,7 +486,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.6.9"
+ "version": "3.8.16"
}
},
"nbformat": 4,
diff --git a/setup.py b/setup.py
index e67e2ffaed5470af644c2b63895fe749fd0e6a58..93d8c0b28b7eeb74aeeed091d4c73d4138758a49 100644
--- a/setup.py
+++ b/setup.py
@@ -6,5 +6,6 @@ setup(
version='0.2.2',
install_requires=[
'tvb-library',
+ 'pooch',
]
)
diff --git a/tvb_ebrains_data/__init__.py b/tvb_ebrains_data/__init__.py
index fdbce0343b566dc34733a40da5f53e9194b32606..437098fe26e6f43dd2905fd1da7481bb7849c0bc 100644
--- a/tvb_ebrains_data/__init__.py
+++ b/tvb_ebrains_data/__init__.py
@@ -1,4 +1,5 @@
from .data import HCPDataset
+from .data import HCPDomhof
from .data import Brains1000Dataset
diff --git a/tvb_ebrains_data/data.py b/tvb_ebrains_data/data.py
index e599f3c08454de1537a04dde46bcd7cfb826dbfc..740d5c889a420b9848585cab83a3e2c67410302f 100644
--- a/tvb_ebrains_data/data.py
+++ b/tvb_ebrains_data/data.py
@@ -3,6 +3,8 @@ import os
import io
import requests
import numpy as np
+import pooch
+from glob import glob
from tvb.simulator.lab import *
logger = logging.getLogger(__name__)
@@ -132,6 +134,108 @@ class HCPDataset(DataProxyConnectivityDataset):
return W, D
+class HCPDomhof():
+ """
+ - https://doi.org/10.25493/NVS8-XS5
+ - https://doi.org/10.25493/F9DP-WCQ
+
+ Available parcellations:
+ 031-MIST 038-CraddockSCorr2Level
+ 048-HarvardOxfordMaxProbThr0 056-CraddockSCorr2Level
+ 056-MIST 070-DesikanKilliany
+ 079-Shen2013 086-EconomoKoskinas
+ 092-AALV2 096-HarvardOxfordMaxProbThr0
+ 100-Schaefer17Networks 103-MIST
+ 108-CraddockSCorr2Level 150-Destrieux
+ 156-Shen2013 160-CraddockSCorr2Level
+ 167-MIST 200-Schaefer17Networks
+ 210-Brainnetome 294-Julich-Brain
+
+ The tsv files for the time-series are double space separated and have a trailing space ??
+ ```
+ find . -name '*.tsv' -exec sed -i 's/[ \t]*$//' {} \;
+ find . -name '*.tsv' -exec sed -i 's/ /,/g' {} \;
+ ```
+ """
+ parcellations = {
+ '031-MIST' : (None, None), # connectivity, bold
+ '038-CraddockSCorr2Level' : (None, None),
+ '048-HarvardOxfordMaxProbThr0' : (None, None),
+ '056-CraddockSCorr2Level' : (None, None),
+ '056-MIST' : (None, None),
+ '070-DesikanKilliany' : ('419c59ab4e01059265cad42e5e68d7e58b0381bc27fe47ad4d4218358ba76280',
+ '26812c39d3963924d3c8fdf65dc3e08b6be13e7d6a5f91e7a13734714958f5fc'),
+ '079-Shen2013' : (None, None),
+ '086-EconomoKoskinas' : (None, None),
+ '092-AALV2' : (None, None),
+ '096-HarvardOxfordMaxProbThr0' : (None, None),
+ '100-Schaefer17Networks' : ('de583e85dd4aa1c0521d61d73f290ad6da3ce4cd5b3538c62b7630f03e438157',
+ '650d5bf9a103299c6505a129051b5bc54a413a9d5ddffb574a4d165c12b1457f'),
+ '103-MIST' : (None, None),
+ '108-CraddockSCorr2Level' : (None, None),
+ '150-Destrieux' : (None, None),
+ '156-Shen2013' : (None, None),
+ '160-CraddockSCorr2Level' : (None, None),
+ '167-MIST' : (None, None),
+ '200-Schaefer17Networks' : ('5086f4b3405acff84ffe132cee17c67a90000a3fae98da50d4e14fb55d7f5d57',
+ 'md5:1f25b912465fe651f5338a7f106f5fe0'),
+ '210-Brainnetome' : (None, None),
+ '294-Julich-Brain' : (None, None),
+ }
+
+ def __init__(self, data_root, parcellation='200-Schaefer17Networks'):
+ self.data_root = data_root
+ assert parcellation in self.parcellations, 'Incorrect parcellation name.'
+ _ = pooch.retrieve(
+ url=f'https://object.cscs.ch/v1/AUTH_227176556f3c4bb38df9feea4b91200c/hbp-d000067_Atlas_based_HCP_BOLD_pub/v1.0/{parcellation}.zip',
+ known_hash=self.parcellations[parcellation][1],
+ path=os.path.join(self.data_root, 'bold'),
+ processor=pooch.Unzip(extract_dir='.')
+ )
+ _ = pooch.retrieve(
+ url=f'https://object.cscs.ch/v1/AUTH_227176556f3c4bb38df9feea4b91200c/hbp-d000059_Atlas_based_HCP_connectomes_v1.1_pub/{parcellation}.zip',
+ known_hash=self.parcellations[parcellation][0],
+ path=os.path.join(self.data_root, 'connectivity'),
+ processor=pooch.Unzip(extract_dir='.')
+ )
+ self.subjects = [os.path.basename(p) for p in glob(os.path.join(self.data_root, 'bold', parcellation, '*[0-9]')) ]
+ self.parcellation = parcellation
+
+ def load_sc(self, subj):
+ """
+ Load structural connectivity.
+
+ Parameters
+ ----------
+ subj : str
+ Subject id from `self.subjects`.
+
+ Returns
+ -------
+ weights: ndarray
+ Weights matrix [N,N].
+ tract_length: ndarray
+ Tract length matrix [N,N].
+ """
+ weights = np.loadtxt(
+ os.path.join(self.data_root, 'connectivity', self.parcellation, '1StructuralConnectivity', subj, 'Counts.csv'),
+ delimiter=' ',
+ )
+ tract_lengths = np.loadtxt(
+ os.path.join(self.data_root, 'connectivity', self.parcellation, '1StructuralConnectivity', subj, 'Lengths.csv'),
+ delimiter=' ',
+ )
+ return weights, tract_lengths
+
+ def load_bold(self, subj):
+ bold = np.loadtxt(
+ os.path.join(self.data_root, 'bold', self.parcellation, subj, 'rfMRI_REST1_LR_BOLD.tsv'),
+ )
+ return bold
+
+
+
+
class Brains1000Dataset(DataProxyConnectivityDataset):
"""