Skip to content

bioinfodlsu/rispice

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

RiSPICE

RiSPICE (Rice SNP Prioritization Integrating Chromatin Effects) is a method for prioritizing non-coding rice variants by predicting their impact on chromatin features using a fine-tuned DNA language model.

This repository accompanies the paper:

Prioritizing Non-coding Variants in Rice GWAS Loci with a Chromatin-Informed DNA Language Model

It includes:

  • Data preprocessing and dataset generation
  • Fine-tuning DNABERT-2 with LoRA
  • Model evaluation
  • Variant effect scoring
  • Downstream case study analyses

Overview

RiSPICE predicts chromatin feature changes caused by non-coding variants and ranks variants according to their predicted regulatory impact. The repository includes:

  • Data preprocessing and dataset generation
  • Fine-tuning DNABERT-2 with LoRA
  • Model evaluation
  • Variant effect scoring
  • Downstream case study analyses

Models & Adapters

RiSPICE builds upon DNABERT-2-117M. We provide the RiSPICE Base model, which contains a trained multi-label chromatin feature prediction head, together with LoRA adapters optimized for different input sequence lengths.

Download the RiSPICE Base model together with the desired LoRA adapter before running inference.

hf download <repo_id> --local-dir <local_directory>

Citation

If you use RiSPICE in your research, please cite the associated paper.

License

This project is released under the MIT License. See the LICENSE file for details.

Contributors

Languages