https://github.com/genentech/reglm

Toolkit for training hyenaDNA-based autoregressive language models on DNA sequences.
https://github.com/genentech/reglm

Last synced: about 1 year ago
JSON representation

Toolkit for training hyenaDNA-based autoregressive language models on DNA sequences.

Host: GitHub
URL: https://github.com/genentech/reglm
Owner: Genentech
License: mit
Created: 2023-10-02T02:39:49.000Z (over 2 years ago)
Default Branch: main
Last Pushed: 2024-06-17T17:34:14.000Z (almost 2 years ago)
Last Synced: 2024-09-26T02:01:41.067Z (over 1 year ago)
Language: Jupyter Notebook
Homepage:
Size: 4.14 MB
Stars: 24
Watchers: 7
Forks: 0
Open Issues: 0
Metadata Files:
- Readme: README.md
- License: LICENSE.txt
- Authors: AUTHORS.md

README

# regLM
regLM is a toolkit for training hyenaDNA-based autoregressive language models on DNA sequences and generating novel regulatory elements.

![regLM schematic](fig1.png)

## Documentation

[Documentation](https://genentech.github.io/regLM)

## Tutorials

[Tutorials](tutorials)

## Installation

### 1. Install HyenaDNA
To use regLM, first install HyenaDNA from GitHub following the instructions: https://github.com/HazyResearch/hyena-dna

### 2. Install regLM
```
git clone https://github.com/Genentech/regLM.git
cd regLM
pip install .
```

## Publication

https://genome.cshlp.org/content/early/2024/09/24/gr.279142.124.abstract

Lal, A., Garfield, D., Biancalani, T., & Eraslan, G. (2024). Designing realistic regulatory DNA with autoregressive language models. Genome Research.