Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -1,73 +1,57 @@
|
|
| 1 |
---
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
| 4 |
tags:
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
- dilated-cnn
|
| 16 |
-
pretty_name: Dilated CNN Macaque Basal Ganglia model
|
| 17 |
---
|
| 18 |
|
| 19 |
# AllenInstitute/BG-macaque-dilatedCNN
|
| 20 |
|
| 21 |
-
Dilated CNN trained on macaque (*Macaca mulatta*) basal ganglia chromatin accessibility
|
| 22 |
|
| 23 |
-
##
|
| 24 |
-
|
| 25 |
-
| Field | Value |
|
| 26 |
-
|---|---|
|
| 27 |
-
| Architecture | Dilated CNN |
|
| 28 |
-
| Species | *Macaca mulatta* (macaque) |
|
| 29 |
-
| Brain region | Basal ganglia |
|
| 30 |
-
| Number of tracks | 58 |
|
| 31 |
-
| Framework | PyTorch Lightning |
|
| 32 |
-
| Checkpoint format | `.ckpt` |
|
| 33 |
|
| 34 |
-
##
|
|
|
|
| 35 |
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
For broader context on the associated multiomic studies and the consensus cell type taxonomy used to define tracks, see the [BG Cell Atlas sequence modeling repository](https://huggingface.co/AllenInstitute/BG-cell-atlas-sequence-modeling).
|
| 39 |
|
| 40 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
|
| 42 |
-
|
| 43 |
|
| 44 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
|
| 46 |
## Usage
|
| 47 |
|
| 48 |
```python
|
| 49 |
-
from
|
| 50 |
-
|
| 51 |
-
model = load_model("AllenInstitute/BG-macaque-dilatedCNN")
|
| 52 |
-
```
|
| 53 |
-
|
| 54 |
-
## Target Tracks
|
| 55 |
-
|
| 56 |
-
The full list of 58 tracks is accessible programmatically after loading:
|
| 57 |
|
| 58 |
-
```python
|
| 59 |
model = load_model("AllenInstitute/BG-macaque-dilatedCNN")
|
| 60 |
-
print(model.target_names)
|
| 61 |
```
|
| 62 |
-
|
| 63 |
-
## Citation
|
| 64 |
-
|
| 65 |
-
```bibtex
|
| 66 |
-
@article{Johansen2025,
|
| 67 |
-
title = {Cross-species consensus atlas of the primate basal ganglia},
|
| 68 |
-
author = {Johansen, Nelson and Fu, Yuanyuan and others},
|
| 69 |
-
journal = {bioRxiv},
|
| 70 |
-
year = {2025},
|
| 71 |
-
doi = {10.64898/2025.12.15.694496}
|
| 72 |
-
}
|
| 73 |
-
```
|
|
|
|
| 1 |
---
|
| 2 |
+
library_name: regrex
|
| 3 |
+
license: other
|
| 4 |
+
license_name: allen-institute-software-license
|
| 5 |
+
license_link: LICENSE
|
| 6 |
tags:
|
| 7 |
+
- pytorch
|
| 8 |
+
- lightning
|
| 9 |
+
- genomics
|
| 10 |
+
- regrex
|
| 11 |
+
- neuroscience
|
| 12 |
+
- regulatory-genomics
|
| 13 |
+
- sequence-to-function
|
| 14 |
+
- basal-ganglia
|
| 15 |
+
- macaque
|
| 16 |
+
- dilated_cnn
|
|
|
|
|
|
|
| 17 |
---
|
| 18 |
|
| 19 |
# AllenInstitute/BG-macaque-dilatedCNN
|
| 20 |
|
| 21 |
+
Dilated CNN trained on macaque (*Macaca mulatta*) basal ganglia chromatin accessibility across 58 cell-type tracks. Part of the DNA sequence-modeling resources for the basal ganglia cell atlas. Published in [Johansen, Fu et al., 2025](https://doi.org/10.64898/2025.12.15.694496).
|
| 22 |
|
| 23 |
+
### Background
|
| 24 |
+
Learns cis-regulatory sequence logic underlying basal ganglia cell-type specialization from snATAC-seq accessibility profiles. Use it to score candidate enhancers, run in silico mutagenesis, or guide cell-type-specific sequence design.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
|
| 26 |
+
### Availability
|
| 27 |
+
Currently private. RegRex (formerly `enhancer-designer`) is an internal Allen Institute tool with a public release planned. For early access contact Kasia Kedzierska (kasia.kedzierska@alleninstitute.org).
|
| 28 |
|
| 29 |
+
## Model Details
|
|
|
|
|
|
|
| 30 |
|
| 31 |
+
- **Architecture:** dilated_cnn
|
| 32 |
+
- **Species:** macaque
|
| 33 |
+
- **Number of tracks:** 58
|
| 34 |
+
- **Framework:** PyTorch Lightning
|
| 35 |
+
- **Checkpoint format:** Lightning .ckpt
|
| 36 |
|
| 37 |
+
## Target Tracks
|
| 38 |
|
| 39 |
+
- AMY-SLEA-BNST_D1_GABA
|
| 40 |
+
- AMY-SLEA-BNST_GABA
|
| 41 |
+
- Astrocyte
|
| 42 |
+
- BAM
|
| 43 |
+
- BF_SKOR1_Glut
|
| 44 |
+
- COP
|
| 45 |
+
- Endo
|
| 46 |
+
- Ependymal
|
| 47 |
+
- GPe-NDB-SI_LHX6-LHX8-GBX1_GABA
|
| 48 |
+
- GPe_MEIS2-SOX6_GABA
|
| 49 |
+
- ... (48 more)
|
| 50 |
|
| 51 |
## Usage
|
| 52 |
|
| 53 |
```python
|
| 54 |
+
from regrex.models import load_model
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
|
|
|
|
| 56 |
model = load_model("AllenInstitute/BG-macaque-dilatedCNN")
|
|
|
|
| 57 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|