https://github.com/cbica/ccl_nmf_prediction
Prediction tool for 7 CCL_NMF component loadings
Science Score: 26.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (13.1%) to scientific vocabulary
Repository
Prediction tool for 7 CCL_NMF component loadings
Statistics
- Stars: 0
- Watchers: 1
- Forks: 1
- Open Issues: 1
- Releases: 0
Metadata Files
README.md
CCL-NMF loading estimation from structural MRI data and basic demographics
This repository provides a lightweight and practical solution for estimating Coupled Cross-sectional and Longitudinal Non-negative Matrix Factorization (CCL-NMF) loading coefficients in new datasets, without the need to re-run the full CCL-NMF pipeline. We provide pre-trained regression models that approximate the subject-level cross-sectional CCL-NMF loading coefficients using regional brain volumes derived from T1-weighted MRI and basic demographic information (age, sex, intracranial volume) to make CCL-NMF loadings more accessible for downstream research. The regional volumes used in these models are not harmonized. This design choice eliminates the need for users to perform data harmonization, which is often impractical in small or single-site datasets.
Installation
It is strongly recommended to install the application in a new virtual environment, such as via Miniconda or venv. Please see the respective pages to learn how to set up a virtual environment for python 3.10 on your system and activate it.
When your desired environment is active:
git clone https://github.com/CBICA/CCL_NMF_Prediction.git
cd CCL_NMF_Prediction
pip install -e .
Usage
CCLNMFPrediction requires 3 arguments:
- -i : The input DLMUSE ROI CSV. This is a comma-separated tabular data containing volumes for ROIs with columns matching the DLMUSE ROI indices. This should be the output of the NiChart_DLMUSE tool.
- -d : The demographics CSV containing at least MRID (string, the subject identifier), Age (integer), Sex ("M"/"F").
- -o : The path to the output CSV file containing CCL-NMF component predictions. Must be writable.
An example command is:
ccl_nmf_prediction -i /path/to/input.csv -d /path/to/demographics.csv -o /path/to/output.csv
Testing
Please see test for testing and replication
Owner
- Name: Center for Biomedical Image Computing & Analytics (CBICA)
- Login: CBICA
- Kind: organization
- Email: software@cbica.upenn.edu
- Location: Philadelphia, PA
- Website: https://www.med.upenn.edu/cbica/
- Twitter: CBICAannounce
- Repositories: 21
- Profile: https://github.com/CBICA
CBICA focuses on the development and application of advanced computation techniques.
GitHub Events
Total
- Push event: 4
- Pull request event: 4
- Fork event: 1
Last Year
- Push event: 4
- Pull request event: 4
- Fork event: 1
Dependencies
- numpy *
- pandas *
- scikit-learn *