Overview
This is the CellHermes model, based on the LLaMA-3.1-8B-instruct architecture developed by Meta, fine-tuned using single-cell RNA sequencing (scRNA-seq) datasets from CellxGene and PPI network from BioGRID. CellHermes is an innovative framework for adapting existing large language models (LLMs) to omics data by transforming various omics data into natural language. This transformation enables LLMs to leverage their powerful understanding and reasoning capabilities for various omics tasks, including perturbation prediction, cell fitness prediction, gene-disease association classification, etc.
Training Data
This model was trained on over 0.2 million human transcriptome instruction data and 0.2 million PPI network datasets from CellxGene and BioGRID. This dataset covers a broad range of cell types and conditions from multiple tissues in both human.
Tasks
This model is designed for:
- Omics data understanding.
CellHermes Links
- GitHub: https://github.com/theislab/CellHermes
- Paper:
LLaMa-3.1-8B-Instruct Links
- Paper: https://arxiv.org/abs/2407.21783
- Hugging Face: https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct