Science Score: 31.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
○.zenodo.json file
-
○DOI references
-
○Academic publication links
-
○Committers with academic emails
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (11.7%) to scientific vocabulary
Repository
Basic Info
- Host: GitHub
- Owner: cossio
- License: other
- Language: Python
- Default Branch: diffRBM
- Size: 36.1 MB
Statistics
- Stars: 1
- Watchers: 3
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
DiffRBM Python package
Python module to train a Differential RBM (DiffRBM), as described in the paper
Barbara Bravi, Andrea Di Gioacchino, Jorge Fernandez-de-Cossio-Diaz, Aleksandra M Walczak, Thierry Mora, Simona Cocco, Rémi Monasson (2023) A transfer-learning approach to predict antigen immunogenicity and T-cell receptor specificity eLife 12:e85126.
Usage of this repo or modifications derived from it should cite the above publication. You can also use the included CITATION.bib file.
This code is based on a fork of https://github.com/jertubiana/PGM, which implements training of a Restricted Boltzmann machine.
On top of that, we implemented here the class DiffRBM (in the file source/diffrbm.py). The following code gives an example of how to construct it:
``` RBM_back # pre-trained background RBM
....
construct the full RBM (back + diff units)
RBMpost = rbm.RBM( visible = RBMback.visible, # nature of visible units hidden = RBMback.hidden, # nature of hidden units nv = RBMback.nv, # number of visible units ncv = RBMback.ncv, # number of states nh = RBMback.nh + diffnh # hidden units = background hidden units + diffRBM units )
construct a DiffRBM object
dRBM = diffrbm.DiffRBM(RBMback, RBMpost)
ensure parameters of the post and back models are in-sync
dRBM.updatepostfrom_back(vlayer=True, hlayer=True) ```
At this point dRBM has been initialized. The background RBM parameters have been copied from the pre-trained RBM_back. Now we need to train the diffRBM units on "selected data". To do this, use the fit_top function,
dRBM.fit_top(sel_data)
See the example notebooks in https://github.com/bravib/diffRBMimmunogenicityTCRspecificity for more details.
Requirements
Python (tested with v3.9), Numpy (tested with v1.23), Numba (tested with v0.56). See also the Requirements section in https://github.com/jertubiana/PGM.
Citation
If you use this code, please cite:
@article{bravi2022learning,
title={Learning the differences: a transfer-learning approach to predict antigen immunogenicity and T-cell receptor specificity},
author={Bravi, Barbara and Di Gioacchino, Andrea and Fernandez-de-Cossio-Diaz, Jorge and Walczak, Aleksandra M and Mora, Thierry and Cocco, Simona and Monasson, R{\'e}mi},
journal={bioRxiv},
pages={2022--12},
year={2022},
publisher={Cold Spring Harbor Laboratory}
}
Owner
- Name: Jorge Fernandez-de-Cossio-Diaz
- Login: cossio
- Kind: user
- Repositories: 24
- Profile: https://github.com/cossio
Citation (CITATION.bib)
@article {10.7554/eLife.85126,
article_type = {journal},
title = {A transfer-learning approach to predict antigen immunogenicity and T-cell receptor specificity},
author = {Bravi, Barbara and Di Gioacchino, Andrea and Fernandez-de-Cossio-Diaz, Jorge and Walczak, Aleksandra M and Mora, Thierry and Cocco, Simona and Monasson, Rémi},
editor = {Bitbol, Anne-Florence and Eisen, Michael B},
volume = 12,
year = 2023,
month = {sep},
pub_date = {2023-09-08},
pages = {e85126},
citation = {eLife 2023;12:e85126},
doi = {10.7554/eLife.85126},
url = {https://doi.org/10.7554/eLife.85126},
abstract = {Antigen immunogenicity and the specificity of binding of T-cell receptors to antigens are key properties underlying effective immune responses. Here we propose diffRBM, an approach based on transfer learning and Restricted Boltzmann Machines, to build sequence-based predictive models of these properties. DiffRBM is designed to learn the distinctive patterns in amino-acid composition that, on the one hand, underlie the antigen’s probability of triggering a response, and on the other hand the T-cell receptor’s ability to bind to a given antigen. We show that the patterns learnt by diffRBM allow us to predict putative contact sites of the antigen-receptor complex. We also discriminate immunogenic and non-immunogenic antigens, antigen-specific and generic receptors, reaching performances that compare favorably to existing sequence-based predictors of antigen immunogenicity and T-cell receptor specificity.},
keywords = {machine learning, immune response, immunogenicity},
journal = {eLife},
issn = {2050-084X},
publisher = {eLife Sciences Publications, Ltd},
}
GitHub Events
Total
Last Year
Committers
Last synced: over 1 year ago
Top Committers
| Name | Commits | |
|---|---|---|
| cossio | j****z@g****m | 29 |
| jertubiana | j****a@g****m | 19 |
| Jorge FdCD | c****o | 3 |
Issues and Pull Requests
Last synced: over 1 year ago
All Time
- Total issues: 0
- Total pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Total issue authors: 0
- Total pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 0
- Pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Issue authors: 0
- Pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0