Python package to analyze Allen Human Brain genes expression
Navigate to directory where you want to save this package and clone this repository:
git clone https://github.com/roshchupkin/AllenBrainPy.git
Make two directories in your project folder, one for MRI data and for expression data.
In config.py script change DATA_DIR to path to expression directory and MRI_DIR to path to MRI directory.
You can download data from Allen Brain Institute
or from this dropbox link.
Save all csv table in separate folder for every donor in your expression directory.
Original MRI T1 images were transformed to template MNI space. You can download find all transformed image
and template from this link. Save all of them in your MRI directory.
You can update to the newest version using git. Navigate to your folder (where you cloned git repository):
git pull
Python script to analyze VBM results and Allen Human Brain Atlas of gene
expression
optional arguments:
-h, --help show this help message and exit
-o O path to save result folder
-model {cluster_expression,correlation}
Analysis models
-i I path input nifti image of VBM result map
-d {all,caucasian} choose all donors or only caucasian
(choices=['all','caucasian'])
-donor DONOR [DONOR ...]
choose donor(s)
-threshold THRESHOLD value threshold to form clusters from VBM result map
-cl_size_threshold CL_SIZE_THRESHOLD
cluster size threshold to form clusters from VBM
result map
-dist_threshold DIST_THRESHOLD
threshold for distance in voxels to link sample to
clusters
-map_type {p-value,t-stat}
Type of VBM result map. For p-value clusters will be
formed for voxels < threshold,for t-stat clusters will
be formed for voxels > threshold, therefore first
split your image to negative and positive t-stat,save
as two maps with abs values and then run analysis
separately for both images.You can easily extrapolate
these two map types to any other result map.
-result_name RESULT_NAME
name for saving results
-probe_mode {all,best,mean}
gene name for expression analysis
-plot plot boxplot of gene expression
-gene_names GENE_NAMES [GENE_NAMES ...]
gene name for expression analysis
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent