The neuromaps toolbox is designed to help researchers make easy, statistically-rigorous comparisons between brain maps.
Code: https://github.com/netneurolab/neuromaps
Documentation: https://netneurolab.github.io/neuromaps/

---------------------------------------------------------
SYSTEM REQUIREMENTS
---------------------------------------------------------

Currently,  neuromaps works with Python 3.7+ and requires:
- matplotlib
- nibabel (>=3.0)
- nilearn (>=0.7)
- numpy (>=1.14)
- scikit-learn (>=0.17)
- scipy
You will also need to have the Connectome Workbench available on your path in order to use most of the transformation/resampling functionality of neuromaps.
The Connectome Workbench can be found here: https://www.humanconnectome.org/software/connectome-workbench

Neuromaps has been tested on the package versions listed above, and on three different operating systems (Linux, OS, Windows).

---------------------------------------------------------
INSTALLATION GUIDE
---------------------------------------------------------

Neuromaps can be installed through Pypi using `pip install neuromaps` or directly from github:
```
git clone https://github.com/netneurolab/neuromaps
cd neuromaps
pip install .
```
Installation time is minimal.

---------------------------------------------------------
DEMO
---------------------------------------------------------

Please follow the User Guide in the online documentation (https://netneurolab.github.io/neuromaps/usage.html) for a compelte demo of how to use neuromaps.
A brief summary of how to run the key functionalities can be found here.

---Brain maps available in neuromaps:

```
from neuromaps.datasets import available_annotations
for annotation in available_annotations():
    print(annotation)
```

---Fetch a brain map:

```
from neuromaps.datasets import fetch_annotation
annotation = fetch_annotation(source='neurosynth')
```

---Transforming brain maps to the same space:

```
from neuromaps.datasets import fetch_annotation
from neuromaps.resampling import resample_images

neurosynth = fetch_annotation(source='neurosynth')
abagen = fetch_annotation(source='abagen')
ns_res, aba_res = resample_images(src=neurosynth, trg=abagen,
                                  src_space='MNI152', trg_space='fsaverage',
                                  method='linear', resampling='downsample_only')
```

---Generating spatial nulls and comparing brain maps:

```
from neuromaps import datasets, images, nulls, resampling, stats

neurosynth = datasets.fetch_annotation(source='neurosynth')
abagen = datasets.fetch_annotation(source='abagen')
neurosynth, abagen = resampling.resample_images(neurosynth, abagen, 'MNI152', 'fsaverage')
rotated = nulls.alexander_bloch(neurosynth, atlas='fsaverage', density='10k',
                                n_perm=100, seed=1234)
corr, pval = stats.compare_images(neurosynth, abagen, nulls=rotated)
```

Functionalities in neuromaps take minimal time _except_ generating nulls (depending on the number of rotations).

---------------------------------------------------------
INSTRUCTIONS FOR USE
---------------------------------------------------------

Complete instructions for use of neuromaps can be found on the online documentation (https://netneurolab.github.io/neuromaps/).